Automated design of DSP array processor chips

Details are presented of the DAC (DSP ASIC Compiler) silicon compiler framework. DAC allows a non-specialist to automatically design DSP ASICs and DSP ASIC cores directly form a high level specification. Typical designs take only several minutes and the resulting layouts are comparable in area and performance to handcrafted designs.

A floating point CORDIC based SVD processor

An SVD processor system is presented in which each processing element is implemented using a simple CORDIC unit. The internal recursive loop within the CORDIC module is exploited, with pipelining being used to multiplex the two independent micro-rotations onto a single CORDIC processor. This leads to a high performance and efficient hardware architecture. In addition, a novel method for scale factor correction is presented which only need be applied once at the end of the computation. This also reduces the computation time. The net result is an SVD architecture based on a conventional CORDIC approach, which combines high performance with high silicon area efficiency.

Metodologia para adaptação de microarquiteturas microprogramadas soft-core à uma ISA padrão: estudo do impacto sobre a complexidade de hardware para o padrão MIPS

In academia, it is common to create didactic processors, facing practical disciplines in the area of Hardware Computer and can be used as subjects in software platforms, operating systems and compilers. Often, these processors are described without ISA standard, which requires the creation of compilers and other basic software to provide the hardware / software interface and hinder their integration with other processors and devices. Using reconfigurable devices described in a HDL language allows the creation or modification of any microarchitecture component, leading to alteration of the functional units of data path processor as well as the state machine that implements the control unit even as new needs arise. In particular, processors RISP enable modification of machine instructions, allowing entering or modifying instructions, and may even adapt to a new architecture. This work, as the object of study addressing educational soft-core processors described in VHDL, from a proposed methodology and its application on two processors with different complexity levels, shows that it s possible to tailor processors for a standard ISA without causing an increase in the level hardware complexity, ie without significant increase in chip area, while its level of performance in the application execution remains unchanged or is enhanced. The implementations also allow us to say that besides being possible to replace the architecture of a processor without changing its organization, RISP processor can switch between different instruction sets, which can be expanded to toggle between different ISAs, allowing a single processor become adaptive hybrid architecture, which can be used in embedded systems and heterogeneous multiprocessor environments

Proposta e implementação de uma arquitetura reconfigurável híbrida para aplicações baseadas em fluxo de dados

The increase of applications complexity has demanded hardware even more flexible and able to achieve higher performance. Traditional hardware solutions have not been successful in providing these applications constraints. General purpose processors have inherent flexibility, since they perform several tasks, however, they can not reach high performance when compared to application-specific devices. Moreover, since application-specific devices perform only few tasks, they achieve high performance, although they have less flexibility. Reconfigurable architectures emerged as an alternative to traditional approaches and have become an area of rising interest over the last decades. The purpose of this new paradigm is to modify the device s behavior according to the application. Thus, it is possible to balance flexibility and performance and also to attend the applications constraints. This work presents the design and implementation of a coarse grained hybrid reconfigurable architecture to stream-based applications. The architecture, named RoSA, consists of a reconfigurable logic attached to a processor. Its goal is to exploit the instruction level parallelism from intensive data-flow applications to accelerate the application s execution on the reconfigurable logic. The instruction level parallelism extraction is done at compile time, thus, this work also presents an optimization phase to the RoSA architecture to be included in the GCC compiler. To design the architecture, this work also presents a methodology based on hardware reuse of datapaths, named RoSE. RoSE aims to visualize the reconfigurable units through reusability levels, which provides area saving and datapath simplification. The architecture presented was implemented in hardware description language (VHDL). It was validated through simulations and prototyping. To characterize performance analysis some benchmarks were used and they demonstrated a speedup of 11x on the execution of some applications

Projeto e implementação em FPGA de um processador com conjunto de instrução reconfigurável utilizando VHDL

The Reconfigurable Computing is an intermediate solution at the resolution of complex problems, making possible to combine the speed of the hardware with the flexibility of the software. An reconfigurable architecture possess some goals, among these the increase of performance. The use of reconfigurable architectures to increase the performance of systems is a well known technology, specially because of the possibility of implementing certain slow algorithms in the current processors directly in hardware. Amongst the various segments that use reconfigurable architectures the reconfigurable processors deserve a special mention. These processors combine the functions of a microprocessor with a reconfigurable logic and can be adapted after the development process. Reconfigurable Instruction Set Processors (RISP) are a subgroup of the reconfigurable processors, that have as goal the reconfiguration of the instruction set of the processor, involving issues such formats, operands and operations of the instructions. This work possess as main objective the development of a RISP processor, combining the techniques of configuration of the set of executed instructions of the processor during the development, and reconfiguration of itself in execution time. The project and implementation in VHDL of this RISP processor has as intention to prove the applicability and the efficiency of two concepts: to use more than one set of fixed instructions, with only one set active in a given time, and the possibility to create and combine new instructions, in a way that the processor pass to recognize and use them in real time as if these existed in the fixed set of instruction. The creation and combination of instructions is made through a reconfiguration unit, incorporated to the processor. This unit allows the user to send custom instructions to the processor, so that later he can use them as if they were fixed instructions of the processor. In this work can also be found simulations of applications involving fixed and custom instructions and results of the comparisons between these applications in relation to the consumption of power and the time of execution, which confirm the attainment of the goals for which the processor was developed

Compilação de bytecodes Java para um ambiente de arquitetura reconfigurável

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Hybrid Methods for Resource Allocation and Scheduling Problems in Deterministic and Stochastic Environments

This work presents exact, hybrid algorithms for mixed resource Allocation and Scheduling problems; in general terms, those consist into assigning over time finite capacity resources to a set of precedence connected activities. The proposed methods have broad applicability, but are mainly motivated by applications in the field of Embedded System Design. In particular, high-performance embedded computing recently witnessed the shift from single CPU platforms with application-specific accelerators to programmable Multi Processor Systems-on-Chip (MPSoCs). Those allow higher flexibility, real time performance and low energy consumption, but the programmer must be able to effectively exploit the platform parallelism. This raises interest in the development of algorithmic techniques to be embedded in CAD tools; in particular, given a specific application and platform, the objective if to perform optimal allocation of hardware resources and to compute an execution schedule. On this regard, since embedded systems tend to run the same set of applications for their entire lifetime, off-line, exact optimization approaches are particularly appealing. Quite surprisingly, the use of exact algorithms has not been well investigated so far; this is in part motivated by the complexity of integrated allocation and scheduling, setting tough challenges for ``pure'' combinatorial methods. The use of hybrid CP/OR approaches presents the opportunity to exploit mutual advantages of different methods, while compensating for their weaknesses. In this work, we consider in first instance an Allocation and Scheduling problem over the Cell BE processor by Sony, IBM and Toshiba; we propose three different solution methods, leveraging decomposition, cut generation and heuristic guided search. Next, we face Allocation and Scheduling of so-called Conditional Task Graphs, explicitly accounting for branches with outcome not known at design time; we extend the CP scheduling framework to effectively deal with the introduced stochastic elements. Finally, we address Allocation and Scheduling with uncertain, bounded execution times, via conflict based tree search; we introduce a simple and flexible time model to take into account duration variability and provide an efficient conflict detection method. The proposed approaches achieve good results on practical size problem, thus demonstrating the use of exact approaches for system design is feasible. Furthermore, the developed techniques bring significant contributions to combinatorial optimization methods.

Integración del descodificador OpenSVC en el DSP del OMAP3530

Hoy en día el uso de dispositivos portátiles multimedia es ya una realidad totalmente habitual. Además, estos dispositivos tienen una capacidad de cálculo y unos recursos gráficos y de memoria altos, tanto es así que por ejemplo en un móvil se pueden reproducir vídeos de muy alta calidad o tener capacidad para manejar entornos 3D. El precio del uso de estos recursos es un mayor consumo de batería que en ocasiones es demasiado alto y acortan en gran medida la vida de la carga útil de la batería. El Grupo de Diseño Electrónico y Microelectrónico de la Universidad Politécnica de Madrid ha abierto una línea de trabajo que busca la optimización del consumo de energía en este tipo de dispositivos, concretamente en el ámbito de la reproducción de vídeo. El enfoque para afrontar la solución del problema se basa en obtener un mayor rendimiento de la batería a costa de disminuir la experiencia multimedia del usuario. De esta manera, cuando la carga de la batería esté por debajo de un determinado umbral mientras el dispositivo esté reproduciendo un vídeo de alta calidad será el dispositivo quien se autoconfigure dinámicamente para consumir menos potencia en esta tarea, reduciendo la tasa de imágenes por segundo o la resolución del vídeo que se descodifica. Además de lo citado anteriormente se propone dividir la descodificación y la representación del vídeo en dos procesadores, uno de propósito general y otro para procesado digital de señal, con esto se consigue que tener la misma capacidad de cálculo que con un solo procesador pero a una frecuencia menor. Para materializar la propuesta se usará la tarjeta BeagleBoard basada en un procesador multinúcleo OMAP3530 de Texas Instrument que contiene dos núcleos: un ARM1 Cortex-A8 y un DSP2 de la familia C6000. Este procesador multinúcleo además permite modificar la frecuencia de reloj y la tensión de alimentación dinámicamente para conseguir reducir de este modo el consumo del terminal. Por otro lado, como reproductor de vídeos se utilizará una versión de MPlayer que integra un descodificador de vídeo escalable que permite elegir dinámicamente la resolución o las imágenes por segundo que se decodifican para posteriormente mostrarlas. Este reproductor se ejecutará en el núcleo ARM pero debido a la alta carga computacional de la descodificación de vídeos, y que el ARM no está optimizado para este tipo de procesado de datos, el reproductor debe encargar la tarea de la descodificación al DSP. El objetivo de este Proyecto Fin de Carrera consiste en que mientras el descodificador de vídeo está ejecutándose en el núcleo DSP y el Mplayer en el núcleo ARM del OMAP3530 se pueda elegir dinámicamente qué parte del vídeo se descodifica, es decir, seleccionar en tiempo real la calidad o capa del vídeo que se quiere mostrar. Haciendo esto, se podrá quitar carga computacional al núcleo ARM y asignársela al DSP el cuál puede procesarla a menor frecuencia para ahorrar batería. 1 ARM: Es una arquitectura de procesadores de propósito general basada en RISC (Reduced Instruction Set Computer). Es desarrollada por la empresa inglesa ARM holdings. 2 DSP: Procesador Digital de Señal (Digital Signal Processor). Es un sistema basado en procesador, el cual está orientado al cálculo matemático a altas velocidad. Generalmente poseen varias unidades aritmético-lógicas (ALUs) para conseguir realizar varias operaciones simultáneamente. SUMMARY. Nowadays, the use of multimedia devices is a well known reality. In addition, these devices have high graphics and calculus performance and a lot of memory as well. In instance, we can play high quality videos and 3D environments in a mobile phone. That kind of use may increase the device's power consumption and make shorter the battery duration. Electronic and Microelectronic Design Group of Technical University of Madrid has a research line which is looking for optimization of power consumption while these devices are playing videos. The solution of this trouble is based on taking more advantage of battery by decreasing multimedia user experience. On this way, when battery charge is under a threshold while device is playing a high quality video the device is going to configure itself dynamically in order to decrease its power consumption by decreasing frame per second rate, video resolution or increasing the noise in the decoded frame. It is proposed splitting decoding and representation tasks in two processors in order to have the same calculus capability with lower frecuency. The first one is specialized in digital signal processing and the other one is a general purpose processor. In order to materialize this proposal we will use a board called BeagleBoard which is based on a multicore processor called OMAP3530 from Texas Instrument. This processor includes two cores: ARM Cortex-A8 and a TMS320C64+ DSP core. Changing clock frequency and supply voltage is allowed by OMAP3530, we can decrease the power consumption on this way. On the other hand, MPlayer will be used as video player. It includes a scalable video decoder which let us changing dynamically the resolution or frames per second rate of the video in order to show it later. This player will be executed by ARM core but this is not optimized for this task, for that reason, DSP core will be used to decoding video. The target of this final career project is being able to choose which part of the video is decoded each moment while decoder is executed by DSP and Mplayer by ARM. It will be able to change in real time the video quality, resolution and frames per second that user want to show. On this way, reducing the computational charge within the processor will be possible.

Provisión de servicios de procesado de imágenes basados en arquitecturas orientadas a servicios

Con este proyecto se pretende crear un procedimiento general para la implantación de aplicaciones de procesado de imágenes en cámaras de video IP y la distribución de dicha información mediante Arquitecturas Orientadas a Servicios (SOA). El objetivo principal es crear una aplicación que se ejecute en una cámara de video IP y realice un procesado básico sobre las imágenes capturadas (detección de colores, formas y patrones) permitiendo distribuir el resultado del procesado mediante las arquitecturas SOA descritas en la especificación DPWS (Device Profile for Web Services). El estudio se va a centrar principalmente en la transformación automática de código de procesado de imágenes escrito en Matlab (archivos .m) a un código C ANSI (archivos .c) que posteriormente se compilará para la arquitectura del procesador de la cámara (arquitectura CRIS, similar a la RISC pero con un conjunto reducido de instrucciones). ABSTRACT. This project aims to create a general procedure for the implementation of image processing applications in IP video cameras and the distribution of such information through Service Oriented Architectures (SOA). The main goal is to create an application that runs on IP video camera and carry out a basic processing on the captured images ( color detection, shapes and patterns) allowing to distribute the result of process by SOA architectures described in the DPWS specification (Device Profile for Web Services). The study will focus primarily on the automated transform of image processing code written in Matlab files (. M) to ANSI C code files (. C) which is then compiled to the processor architecture of the camera (CRIS architecture , similar to the RISC but with a reduced instruction set).

From rhetoric to practice : issues in teaching and learning touch keyboarding

Touch keyboarding as a vocational skill is disappearing at a time when students and educators across alleducational sectors are expected to use a computer keyboard on a regular basis. there is documentation surrounding the embedding of Information and Communication Technology (ICT) within the curricula and yet within the National Training Packages touch keyboarding, previously considered a core component, is now an elective in the Business Services framework. This situation is an odds with current practice overseas where touch keyboarding is a component of primary and secondary curricula. From Rhetoric to Practice explores the current issues and practice in teaching and learning touch keyboarding in primary, secondary and tertiary institutions. Through structured interview participants detailed current practice of teachers and their students. Further, tertiary students participated in a training program aimed at achquiring touch keyboarding as a skill to enhance their studies. The researcher's background experience of fifteen years teaching touch keyboarding and computer literacty to adults and 30 years in Business Services trade provides a strong basis for this project. The teaching experience is enhanced by industry experience in administration, course coordination in technical, community and tertiary institutions and a strong commitment to the efficient usage of a computer by all. The findings of this project identified coursework expectations requiring all students from kindergarten to tertiary to use a computer keyboard on a weekly basis and that neither teaching nor learning tough keyboarding appears in the primary, secondary and tertiary curricula in New South Wales. Further, teachers recognised tough keyboarding as the prefered style over 'hunt and peck' keyboarding while acknowledging the teaching and learning difficulties of time constraints, the need for qualified touch keyboarding teachers and issues arising when retraining students from existing poor habits. In conclusion, this project recommends that computer keyboarding be defined as a writing tool for education, vocation and life, with early instruction set in primary schooling area and embedding touch keyboarding with the secondary, technical and tertiary areas and finally to draw the attention of educational authorities to the Duty Of Care aspects associated with computer keyboarding in the classroom.

Cohort normalization based sparse representation for undersampled face recognition

Abstract. In recent years, sparse representation based classification(SRC) has received much attention in face recognition with multipletraining samples of each subject. However, it cannot be easily applied toa recognition task with insufficient training samples under uncontrolledenvironments. On the other hand, cohort normalization, as a way of mea-suring the degradation effect under challenging environments in relationto a pool of cohort samples, has been widely used in the area of biometricauthentication. In this paper, for the first time, we introduce cohort nor-malization to SRC-based face recognition with insufficient training sam-ples. Specifically, a user-specific cohort set is selected to normalize theraw residual, which is obtained from comparing the test sample with itssparse representations corresponding to the gallery subject, using poly-nomial regression. Experimental results on AR and FERET databases show that cohort normalization can bring SRC much robustness against various forms of degradation factors for undersampled face recognition.

Comparison of high level FPGA hardware design for solving tri-diagonal linear systems

Reconfigurable computing devices can increase the performance of compute intensive algorithms by implementing application specific co-processor architectures. The power cost for this performance gain is often an order of magnitude less than that of modern CPUs and GPUs. Exploiting the potential of reconfigurable devices such as Field-Programmable Gate Arrays (FPGAs) is typically a complex and tedious hardware engineering task. Re- cently the major FPGA vendors (Altera, and Xilinx) have released their own high-level design tools, which have great potential for rapid development of FPGA based custom accelerators. In this paper, we will evaluate Altera’s OpenCL Software Development Kit, and Xilinx’s Vivado High Level Sythesis tool. These tools will be compared for their per- formance, logic utilisation, and ease of development for the test case of a Tri-diagonal linear system solver.

Synthesis of ASIPs for DSP algorithms

ASICs offer the best realization of DSP algorithms in terms of performance, but the cost is prohibitive, especially when the volumes involved are low. However, if the architecture synthesis trajectory for such algorithms is such that the target architecture can be identified as an interconnection of elementary parameterized computational structures, then it is possible to attain a close match, both in terms of performance and power with respect to an ASIC, for any algorithmic parameters of the given algorithm. Such an architecture is weakly programmable (configurable) and can be viewed as an application specific integrated processor (ASIP). In this work, we present a methodology to synthesize ASIPs for DSP algorithms. (C) 1999 Elsevier Science B.V. All rights reserved.

Minimizing buffer requirements under rate-optimal schedule in regular dataflow networks

Large-grain synchronous dataflow graphs or multi-rate graphs have the distinct feature that the nodes of the dataflow graph fire at different rates. Such multi-rate large-grain dataflow graphs have been widely regarded as a powerful programming model for DSP applications. In this paper we propose a method to minimize buffer storage requirement in constructing rate-optimal compile-time (MBRO) schedules for multi-rate dataflow graphs. We demonstrate that the constraints to minimize buffer storage while executing at the optimal computation rate (i.e. the maximum possible computation rate without storage constraints) can be formulated as a unified linear programming problem in our framework. A novel feature of our method is that in constructing the rate-optimal schedule, it directly minimizes the memory requirement by choosing the schedule time of nodes appropriately. Lastly, a new circular-arc interval graph coloring algorithm has been proposed to further reduce the memory requirement by allowing buffer sharing among the arcs of the multi-rate dataflow graph. We have constructed an experimental testbed which implements our MBRO scheduling algorithm as well as (i) the widely used periodic admissible parallel schedules (also known as block schedules) proposed by Lee and Messerschmitt (IEEE Transactions on Computers, vol. 36, no. 1, 1987, pp. 24-35), (ii) the optimal scheduling buffer allocation (OSBA) algorithm of Ning and Gao (Conference Record of the Twentieth Annual ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages, Charleston, SC, Jan. 10-13, 1993, pp. 29-42), and (iii) the multi-rate software pipelining (MRSP) algorithm (Govindarajan and Gao, in Proceedings of the 1993 International Conference on Application Specific Array Processors, Venice, Italy, Oct. 25-27, 1993, pp. 77-88). Schedules generated for a number of random dataflow graphs and for a set of DSP application programs using the different scheduling methods are compared. The experimental results have demonstrated a significant improvement (10-20%) in buffer requirements for the MBRO schedules compared to the schedules generated by the other three methods, without sacrificing the computation rate. The MBRO method also gives a 20% average improvement in computation rate compared to Lee's Block scheduling method.

Dynamically Reconfigurable Regular Expression Matching Architecture

Regular Expressions are generic representations for a string or a collection of strings. This paper focuses on implementation of a regular expression matching architecture on reconfigurable fabric like FPGA. We present a Nondeterministic Finite Automata based implementation with extended regular expression syntax set compared to previous approaches. We also describe a dynamically reconfigurable generic block that implements the supported regular expression syntax. This enables formation of the regular expression hardware by a simple cascade of generic blocks as well as a possibility for reconfiguring the generic blocks to change the regular expression being matched. Further,we have developed an HDL code generator to obtain the VHDL description of the hardware for any regular expression set. Our optimized regular expression engine achieves a throughput of 2.45 Gbps. Our dynamically reconfigurable regular expression engine achieves a throughput of 0.8 Gbps using 12 FPGA slices per generic block on Xilinx Virtex2Pro FPGA.