Tuesday, March 11, 2008
| 10:00 - 12:00 | Session | Test, Yield and Costs | UB 1.1 | |
| University of Limerick | Ian Grout | CMOS Implementation for Data Converter BIST/BOST | Two silicon implementations for BIST and BOST solutions for data converter self-test are to be demonstrated. In these solutions, self-test functions for 16-18 bit data converter designs are undertaken with communications to an external PC. The systems provide the capability for on-chip or on-DIB signal generation, data capture and communications. | 1 |
| University of Siegen | Michael Wahl | Design Cost Controlling | CostMoS-G is a tool for collaborative work on cost optimal designs. The user can define arbitrary cost models for the products life cycle. Users are designers, managers, and controllers. Monte-Carlo simulation allows dealing with value ranges, and the integrated time model permits cost predictions for future reviews. | 2 |
| University of Bremen | Christian Genz | A graphical SystemC refinement tool | We present a graphical design tool that supports the manual refinement of SystemC models with focus on the following features. Separation of HW/SW partitions encapsulated in distinct hierarchies. Highlighting of synthesizable and non-synthesizable fragments of the design. | 3 |
| University of Bremen | Kishore Duganapalli | GA based ATPG Tool for Crosstalk Induced Logic Faults between On-chip Aggressor and Victim | The ATPG tool proposed here based on Genetic Algorithm, reads the circuit description and generates test patterns for the given set of aggressors and victims in DSM chips. | 4 |
| University of Bremen | Shehzad Hasan | Test Compaction of Crosstalk Faults through Fault List Reordering | Crosstalk noise may cause undesirable effects including excessive overshoot, undershoot, glitches, additional signal delay as well as signal speed-up. An algorithm is proposed for generating test patterns that produce maximal crosstalk effect on any interconnect of a circuit using existing ATPG tool and coupling influence between interconnects. The fault list is then reordered so as to reduce the total number of test patterns. | 5 |
| 12:00 - 14:00 | Session | Bringing Ideas into FPGAs | UB 1.2 | |
| Technical University of Braunschweig | Henning Sahlbach, Sean Whitty | FlexFilm: Real-time digital film processing with a FPGA-based reconfigurable platform | The FlexFilm board is a reconfigurable platform based on four Virtex II Pro FPGAs. Several run-time reconfigurable algorithms for real-time digital film processing for resolutions up to 2048x1556 pixels@24 FPS in 10-Bit RGB are demonstrated. The system achieves a sustained performance of 170 GOPS. | 1 |
| Ulm University | Frank Slomka | Submarine Explorer – A low cost AUV | The demonstrator is a low cost autonomous underwater robot. The computer system controlling the robot is design using advanced hardware/software codesign techniques by using state of the art FPGA SOPC technology implementing time critical parts of the system in hardware and using parallel computational resources by a consequent MPSOC approach. Using just one design environment a distributed embedded real-time system is presented. The system contains 4 FPGA boards connected by CAN and Ethernet links. | 2 |
| University of Erlangen-Nuremberg | Dmitrij Kissler | A High-Speed Dynamic Reconfigurable Multilevel Parallel Architecture | The CoMap project investigates the holistic co-design of a new class of highly parameterizable, massively parallel processor architectures. This architecture type is called weakly programmable processor arrays (WPPAs). Two DSP algorithms were implemented with the help of dynamic reconfiguration on the same WPPA hardware, prototyped on FPGA. | 3 |
| University of Pisa | Luca Fanucci | SHINE: FPGA prototyping of SpaceWire IP cores for High Data Rate and Fault Tolerant Invehicle Networking | An FPGA, hosted by a Nallatech BenNUEY PCI-board, is programmed with an 8-port SpaceWire Router IP core featuring an AMBA AHB interface. The router is controlled by a PC through a PCI/AHB bridge. A user-friendly GUI, which hides a dedicated software layer, provides full control over router features (link speed, router table, etc.) and monitors relevant traffic information. | 4 |
| Eindhoven University of Technology | Akash Kumar | PreMaDonna: Predictable Matching of Demands on Networked Architectures | We demonstrate a methodology to generate multiprocessor systems-on-chip from the high-level description of applications, namely synchronous dataflow graphs. An automated framework is presented to generate system-level designs for Xilinx FPGAs in seconds. Attendees will get an opportunity to experiment with this framework and generate multi-processor designs. | 5 |
| 14:00 - 16:00 | Session | System Exploration and Transaction Level Modelling | UB 1.3 | |
| RWTH Aachen University | Torsten Kempf | An ESL Workbench for early MPSoC Design Space Exploration | One major issue in future MPSoC designs is the combined hardware and software development, often referred to HW/SW co-design. Here, both hardware and software are developed in parallel manner, imposing new design challenges to the system architects.To cope with this issue during MPSoC design we have proposed a higher abstraction level, where an abstract processor simulator called Virtual Processing Unit (VPU) is utilized. | 1 |
| University of Tehran | Mahshid Sedghi | TLM Synthesis Studio | TLM Synthesis Studio is an environment to help system level designers start the design and implementation of their systems from transaction level. It supports OSCI TLM library and SystemC. It also provides the designers with a library of configurable TLM components. | 2 |
| Université de Bretagne Sud | Johann Laurent | SoftExplorer | SoftExplorer is a power/energy estimation tool that is developed into the Low Power Design Group of the LESTER laboratory. This tool allows the programmer to rapidly estimate the power and energy consumption of its application executed (for example a C code) on a processor (for example a TI TMS320C62). SoftExplorer owns several power models of processors. | 3 |
| Osaka University | Ittetsu Taniguchi | SoC Architecture Explorer | Architecture design in IP-based design is difficult because of vast design space. SoC Architecture Explorer explores the design space automatically and a number of architectures are output, which have a trade-off relation between performance and hardware area. | 4 |
| Université de Bretagne Sud | Pierre Bomel | A lightweight and remote partially reconfigurable platform | We present the concept of a networked lightweight and partially reconfigurable platform assisted by a remote bitstreams server. We propose a software and hardware architecture as well as a new data-link level network protocol implementation dedicated to dynamic and partial reconfiguration of FPGAs. These systems target embedded applications with very scarce hardware resources taking advantage of dynamic, specific and optimized architectures. | 5 |
| Université de Bretagne Sud | Philippe Coussy | GAUT – A High-Level Synthesis tool for DSP applications | Starting from a pure C function GAUT generates a potentially pipelined architecture composed of a processing unit, a memory unit, a communication unit with a GALS/LIS interface. The synthesis constraints are the data average throughput, the clock frequency, the FPGA/ASIC target technology and optionally the memory architecture/mapping, the I/O timing diagram. GAUT generates an IEEE-P1076 VHDL file and test benches which is an input for commercial logical synthesis tools like ISE/Foundation from Xilinx, Quartus from Altera or DC from Synopsys. | 6 |
| KU Leuven | Andy Lambrechts | The Coffee framework: COmpiler Framework for Energy-aware Exploration | Modern mobile devices need to be extremely energy efficient. This demo presents a unified optimization and exploration framework, from source level transformation to processor architecture design. The retargetable compiler and simulator framework can map applications to a range of processors and memory configurations, simulate and report detailed performance and energy estimates. An accurate energy modeling approach is used which can help to guide the design process. | 7 |
| 16:00 - 18:00 | Session | Software and Hardware in the System | UB 1.4 | |
| Université de Bretagne Sud | Rachid Dafali | µSPIDER CAD TOOL: CASE STUDY OF NOC IP GENERATION FOR FPGA | This demo presents the µSpider CAD tool for network on chip design under latency and bandwidth constraints and described the different steps of the associated design flow. We show: 1) how the tool can be used to automatically generate a NoC IP compliant with Xilinx EDK tool and 2) a real case implementation of a video application running on a Virtex2pro FPGA, the architecture includes 2 Mblaze, 1 PPC and our NoC IP. | 1 |
| Université de Bretagne Sud | Pierre Bomel | A mixed (hardware and software) rapid prototyping platform. | PALMYRE is set of hardware components and software tools enabling the rapid prototyping of telecom-oriented application. DSPs from Texas Instruments and FPGAs from Xilinx can be freely mixed to build a specific computing and communication topology. The demonstration is a pipeline of two DSPs and an FPGA Viterbi decoder. Thanks to the GALS/LIS approach the system is data regulated and can be stopped and restarted without loss of any data or synchronization between DSPs and FPGA. | 2 |
| Eindhoven University of Technology | Andreas Hansson | CoMPSoC - A Composable and Predictable Multi-Processor System-on-Chip Template | We propose a Composable and Predictable Multi-Processor System-on-Chip (CoMPSoC) platform and mapping methodology. Multiple real-time jobs can run simultaneously. Incremental integration and platform virtualisation are supported. The real-time jobs are analysed and verified independent of one another, using state-of-the-art dataflow techniques. | 3 |
| University of Paderborn | Christopher Pohl | Hardware-in-the-Loop Simulations with Matlab/Simulink/ModelSim for FPGA based designflows | This demonstrator shows the design flow and a working example of our Hardware-in-the-Loop Design Environment for FPGAs. Arbitrary VHDL designs can be automatically integrated into simulation or visualization tools, enabling functional verification and real time monitoring of that hardware design. | 4 |
| University of Paderborn | Thorsten Jungeblut | General Purpose VLIW Processor for Multiband-Multistandard applications | This demonstrator proposes our General Purpose VLIW-Processor for applications on mobile phones. Four arithmetic logical units can process scalar or SIMD instructions in parallel. The system is designed for 300 Mhz and achieves a performance of 1.2 GIPS or 2.4 VOPS. | 5 |
| Darmstadt University of Technology | Kurt Ackermann | Demonstration of a Self-Reconfigurable Video-Processing Frame-grabber | Limitations of logic and RAM resources of FPGAs are bottlenecks for many complex algorithms. Modern FPGAs support dynamic reconfiguration. Partitioning of complex algorithms in smaller sequentially reconfigured and executed units makes smaller FPGA devices suitable. A video processing framegrabber, connected to a high resolution matrix camera, is implemented as a demonstrator for a complex self-reconfigurable design on a Virtex-4 based evaluation board. | 6 |
Wednesday, March 12, 2008
| 10:00 - 12:00 | Session | Mixed Signal Design, Test Solutions and More | UB 2.1 | |
| Fraunhofer-Institut für Techno- und Wirtschaftsmathematik | Jochen Broz | Behavioral Modeling using Analog Insydes | Analog Insydes is a tool for modelling and analyzing analog circuits symbolically. For handling model complexity, it provides the methodology of symbolic approximation which is an innovative model-order reduction technique. The process works under full error control and allows the user to specify tolerance margins for the accuracy of the generated model. | 1 |
| University of Limerick | Ian Grout | CMOS Implementation for Data Converter BIST/BOST | Two silicon implementations for BIST and BOST solutions for data converter self-test are to be demonstrated. In these solutions, self-test functions for 16-18 bit data converter designs are undertaken with communications to an external PC. The systems provide the capability for on-chip or on-DIB signal generation, data capture and communications. | 2 |
| RWTH Aachen University | Torsten Kempf | An ESL Workbench for early MPSoC Design Space Exploration | One major issue in future MPSoC designs is the combined hardware and software development, often referred to HW/SW co-design. Here, both hardware and software are developed in parallel manner, imposing new design challenges to the system architects.To cope with this issue during MPSoC design we have proposed a higher abstraction level, where an abstract processor simulator called Virtual Processing Unit (VPU) is utilized. | 3 |
| University of Siegen | Michael Wahl | Design Cost Controlling | CostMoS-G is a tool for collaborative work on cost optimal designs. The user can define arbitrary cost models for the products life cycle. Users are designers, managers, and controllers. Monte-Carlo simulation allows dealing with value ranges, and the integrated time model permits cost predictions for future reviews. | 4 |
| Technical University of Braunschweig | Henning Sahlbach, Sean Whitty | FlexFilm: Real-time digital film processing with a FPGA-based reconfigurable platform | The FlexFilm board is a reconfigurable platform based on four Virtex II Pro FPGAs. Several run-time reconfigurable algorithms for real-time digital film processing for resolutions up to 2048x1556 pixels@24 FPS in 10-Bit RGB are demonstrated. The system achieves a sustained performance of 170 GOPS. | 5 |
| University of Southampton | Reuben Wilcock | Post Manufacture Variability Improvement Using Configurable Analogue Transistors (CATs) | Analogue design is becoming a bottleneck in DSM chips due to increased process variability. A configurable analogue transistor (CAT) structure is proposed which allows post manufacture adjustment of critical devices. The demonstrator shows live results from a silicon prototype, highlighting the adjustability and improvement offered by this approach. | 6 |
| 12:00 - 14:00 | Session | System Analysis, Simulation and Verification | UB 2.2 | |
| University of Bremen | Christian Genz | A graphical SystemC refinement tool | We present a graphical design tool that supports the manual refinement of SystemC models with focus on the following features. Separation of HW/SW partitions encapsulated in distinct hierarchies. Highlighting of synthesizable and non-synthesizable fragments of the design. | 1 |
| Fraunhofer-Institut für Techno- und Wirtschaftsmathematik | Jochen Broz | Behavioral Modeling using Analog Insydes | Analog Insydes is a tool for modelling and analyzing analog circuits symbolically. For handling model complexity, it provides the methodology of symbolic approximation which is an innovative model-order reduction technique. The process works under full error control and allows the user to specify tolerance margins for the accuracy of the generated model. | 2 |
| Ulm University | Frank Slomka | Submarine Explorer – A low cost AUV | The demonstrator is a low cost autonomous underwater robot. The computer system controlling the robot is design using advanced hardware/software codesign techniques by using state of the art FPGA SOPC technology implementing time critical parts of the system in hardware and using parallel computational resources by a consequent MPSOC approach. Using just one design environment a distributed embedded real-time system is presented. The system contains 4 FPGA boards connected by CAN and Ethernet links. | 3 |
| University of Tokyo | Yoshihisa Kojima | A verification environment for high-level designs based on system dependence graphs | This tool provides a verification environment for high-level design descriptions. Program slicing, static code checking, dynamic simulation and formal equivalence checking based on symbolic simulation are realized on top of our ExSDGs (system dependence graphs integrated with abstract syntax trees). | 4 |
| University of Newcastle upon Tyne | Andrey Mokhov | Workcraft: a static data flow structure editing, visualisation and analysis tool | Reliable high-level modelling constructs are crucial to the design of efficient asynchronous circuits. Workcraft is a tool which aims to provide a common, cross-platform environment. The tool offers a GUI-based framework for visual editing, real-time simulation, animation and extendable analysis features for different SDFS types. The models themselves, as well as the supporting tools, are implemented as plug-ins. | 5 |
| 14:00 - 16:00 | Session | SoC, Platforms and SystemC | UB 2.3 | |
| University of Tehran | Mahshid Sedghi | TLM Synthesis Studio | TLM Synthesis Studio is an environment to help system level designers start the design and implementation of their systems from transaction level. It supports OSCI TLM library and SystemC. It also provides the designers with a library of configurable TLM components. | 1 |
| Eindhoven University of Technology | Sander Stuijk | SDF3 | The SDF3 tool implements a state-of-the art design flow that can map a throughput-constrained application modeled with a Synchronous Dataflow Graph onto a NoC-based MP-SoC while providing timing guarantees. The design flow analyzes the resources requirements of the application, allocates the required resources in the NoC-based MP-SoC, and schedules accesses to shared resources. At every step of the flow, the complete state of the design flow can be outputted in XML. | 2 |
| Université de Bretagne Sud | Rachid Dafali | µSPIDER CAD TOOL: CASE STUDY OF NOC IP GENERATION FOR FPGA | This demo presents the µSpider CAD tool for network on chip design under latency and bandwidth constraints and described the different steps of the associated design flow. We show: 1) how the tool can be used to automatically generate a NoC IP compliant with Xilinx EDK tool and 2) a real case implementation of a video application running on a Virtex2pro FPGA, the architecture includes 2 Mblaze, 1 PPC and our NoC IP. | 3 |
| Université de Bretagne Sud | Johann Laurent | SoftExplorer | SoftExplorer is a power/energy estimation tool that is developed into the Low Power Design Group of the LESTER laboratory. This tool allows the programmer to rapidly estimate the power and energy consumption of its application executed (for example a C code) on a processor (for example a TI TMS320C62). SoftExplorer owns several power models of processors. | 4 |
| Osaka University | Ittetsu Taniguchi | SoC Architecture Explorer | Architecture design in IP-based design is difficult because of vast design space. SoC Architecture Explorer explores the design space automatically and a number of architectures are output, which have a trade-off relation between performance and hardware area. | 5 |
| University of Bonn | Andreas Raabe | A Reconfiguration Simulation Library For SystemC | This demonstration presents a library for modelling reconfiguration in SystemC combining IP reuse and high-level modelling with reconfiguration. Control statements and techniques that allow safe process controlling in conjunction with SystemC language constructs are presented. A case study proves its applicability. | 6 |
| Chemnitz University of Technology | Uwe Proß | Specification data gathering | The Software gathers information about the system and its behavior on specification level and will check this specification for consistency. It enables Reuse of testcases as well as requirements and cost engineering. Code export to several tools and languages integrates SpecScribe in the design flow. | 7 |
| Universität Karlsruhe | Jürgen Becker | MORPHEUS integrated toolset | The toolset compiles annotated c-code to ARM binaries while replacing the annotated function calls with calls the equivalent configurations on one of the heterogeneous reconfigurable engines of the MORPHEUS SoC. Additionally the toolset synthesizes the necessary control codes for the NoC to feed data to the engines. | 8 |
| 16:00 - 18:00 | Session | Hardware Solutions | UB 2.4 | |
| Université de Bretagne Sud | Pierre Bomel | A mixed (hardware and software) rapid prototyping platform. | PALMYRE is set of hardware components and software tools enabling the rapid prototyping of telecom-oriented application. DSPs from Texas Instruments and FPGAs from Xilinx can be freely mixed to build a specific computing and communication topology. The demonstration is a pipeline of two DSPs and an FPGA Viterbi decoder. Thanks to the GALS/LIS approach the system is data regulated and can be stopped and restarted without loss of any data or synchronization between DSPs and FPGA. | 1 |
| Université de Bretagne Sud | Pierre Bomel | A lightweight and remote partially reconfigurable platform | We present the concept of a networked lightweight and partially reconfigurable platform assisted by a remote bitstreams server. We propose a software and hardware architecture as well as a new data-link level network protocol implementation dedicated to dynamic and partial reconfiguration of FPGAs. These systems target embedded applications with very scarce hardware resources taking advantage of dynamic, specific and optimized architectures. | 2 |
| Universidad Complutense de Madrid | Francisco Vallejos | OS based Wireless Body Area Network for ECG | We present a complete architecture of a Wireless Body Area Network for ECG (electrocardiogram) monitoring.The network is composed of five nodes and a base station. The nodes proccess the ECG signal that they senseand they send information to the base station when a special event occurs, depending on the application. Wehave developed a new TDMA protocol that reduces the energy consumption associated with the radio. | 3 |
| Université de Bretagne Sud | Philippe Coussy | GAUT – A High-Level Synthesis tool for DSP applications | Starting from a pure C function GAUT generates a potentially pipelined architecture composed of a processing unit, a memory unit, a communication unit with a GALS/LIS interface. The synthesis constraints are the data average throughput, the clock frequency, the FPGA/ASIC target technology and optionally the memory architecture/mapping, the I/O timing diagram. GAUT generates an IEEE-P1076 VHDL file and test benches which is an input for commercial logical synthesis tools like ISE/Foundation from Xilinx, Quartus from Altera or DC from Synopsys. | 4 |
| Eindhoven University of Technology | Andreas Hansson | CoMPSoC - A Composable and Predictable Multi-Processor System-on-Chip Template | We propose a Composable and Predictable Multi-Processor System-on-Chip (CoMPSoC) platform and mapping methodology. Multiple real-time jobs can run simultaneously. Incremental integration and platform virtualisation are supported. The real-time jobs are analysed and verified independent of one another, using state-of-the-art dataflow techniques. | 5 |
| University of Bonn | Stefan Hochgürtel | A broadband FFT-Spectrometer at work | We present a fully functional FFT-spectrometer card, calculating 2048 channels at a bandwidth of up to 1500 Mhz. Multiple parameters as well as the complete spectrometer-core can be altered at run-time over ethernet. Spectra, which are also send to a PC by ethernet, will be presented interactively. | 6 |
| Technical University of Denmark | Jan Madsen | Formal Verification of Design Properties of Hardware Architectures | We present a formal language for hardware models based on the Gezel hardware description language developed and maintained by Virginia Tech, USA. We show how the language can be used in connection with verification by relating the semantical domain to timed-automata using the UPPAAL system. We will also demonstarte a formal verification of design properties of a few simple example circuits including the Simplified Data Encryption Standard (SDES) Algorithm and different algorithmic implementations of the Greatest Coomon Divisor. | 7 |
| University of Rostock | Peter Danielis | Trust-by-Wire in Packet-switched IP Networks: Calling Line Identification Presentation for IP | The packet processing system IPclip (IP Calling Line Identification Presentation) is presented. It is implemented on an FPGA board and configurable at runtime via a graphical configuration tool. The functionality is demonstrated in a localization scenario using an analysis tool and Google Earth. | 8 |
Thursday, March 13, 2008
| 10:00 - 12:00 | Session | System Exploration with Software and Hardware | UB 3.1 | |
| Eindhoven University of Technology | Sander Stuijk | SDF3 | The SDF3 tool implements a state-of-the art design flow that can map a throughput-constrained application modeled with a Synchronous Dataflow Graph onto a NoC-based MP-SoC while providing timing guarantees. The design flow analyzes the resources requirements of the application, allocates the required resources in the NoC-based MP-SoC, and schedules accesses to shared resources. At every step of the flow, the complete state of the design flow can be outputted in XML. | 1 |
| Eindhoven University of Technology | Andreas Hansson | CoMPSoC - A Composable and Predictable Multi-Processor System-on-Chip Template | We propose a Composable and Predictable Multi-Processor System-on-Chip (CoMPSoC) platform and mapping methodology. Multiple real-time jobs can run simultaneously. Incremental integration and platform virtualisation are supported. The real-time jobs are analysed and verified independent of one another, using state-of-the-art dataflow techniques. | 2 |
| University of Bonn | Stefan Hochgürtel | A broadband FFT-Spectrometer at work | We present a fully functional FFT-spectrometer card, calculating 2048 channels at a bandwidth of up to 1500 Mhz. Multiple parameters as well as the complete spectrometer-core can be altered at run-time over ethernet. Spectra, which are also send to a PC by ethernet, will be presented interactively. | 3 |
| University of Paderborn | Christopher Pohl | Hardware-in-the-Loop Simulations with Matlab/Simulink/ModelSim for FPGA based designflows | This demonstrator shows the design flow and a working example of our Hardware-in-the-Loop Design Environment for FPGAs. Arbitrary VHDL designs can be automatically integrated into simulation or visualization tools, enabling functional verification and real time monitoring of that hardware design. | 4 |
| University of Paderborn | Thorsten Jungeblut | General Purpose VLIW Processor for Multiband-Multistandard applications | This demonstrator proposes our General Purpose VLIW-Processor for applications on mobile phones. Four arithmetic logical units can process scalar or SIMD instructions in parallel. The system is designed for 300 Mhz and achieves a performance of 1.2 GIPS or 2.4 VOPS. | 5 |
| Politecnico di Milano | Vittorio Zaccaria | System Tuning Shell: A Design Space Exploration Tool | System Tuning Shell is a design space exploration tool for supporting platform-based design. The tool is a flexible and modular system optimizer that allows users to plug-in custom system models and apply several optimization strategies to derive trade-offs on the system cost functions. | 6 |
| TU München | Christopher Claus | Hardware/software architecture of an algorithm for vision-based real-time vehicle detection in dark environments | The demonstrator runs on an ML310 board from Xilinx. Images are loaded from the connected hard drive into the main memory (DDR SDRAM) which is located on the board. From there, the pixel data can be accessed by a hardware coprocessor called TaillightEngine which is able to detect cars in dark environments. The processed pixels are written back into the main memory from where they can be accessed and be displayed on an external monitor. | 7 |
| 12:00 - 14:00 | Session | High Level Synthesis | UB 3.2 | |
| University of Tehran | Mahshid Sedghi | TLM Synthesis Studio | TLM Synthesis Studio is an environment to help system level designers start the design and implementation of their systems from transaction level. It supports OSCI TLM library and SystemC. It also provides the designers with a library of configurable TLM components. | 1 |
| Université de Bretagne Sud | Philippe Coussy | GAUT – A High-Level Synthesis tool for DSP applications | Starting from a pure C function GAUT generates a potentially pipelined architecture composed of a processing unit, a memory unit, a communication unit with a GALS/LIS interface. The synthesis constraints are the data average throughput, the clock frequency, the FPGA/ASIC target technology and optionally the memory architecture/mapping, the I/O timing diagram. GAUT generates an IEEE-P1076 VHDL file and test benches which is an input for commercial logical synthesis tools like ISE/Foundation from Xilinx, Quartus from Altera or DC from Synopsys. | 2 |
| University of Bremen | Christian Genz | A graphical SystemC refinement tool | We present a graphical design tool that supports the manual refinement of SystemC models with focus on the following features. Separation of HW/SW partitions encapsulated in distinct hierarchies. Highlighting of synthesizable and non-synthesizable fragments of the design. | 3 |
| University of Limerick | Ian Grout | CMOS Implementation for Data Converter BIST/BOST | Two silicon implementations for BIST and BOST solutions for data converter self-test are to be demonstrated. In these solutions, self-test functions for 16-18 bit data converter designs are undertaken with communications to an external PC. The systems provide the capability for on-chip or on-DIB signal generation, data capture and communications. | 4 |
| RWTH Aachen University | Torsten Kempf | An ESL Workbench for early MPSoC Design Space Exploration | One major issue in future MPSoC designs is the combined hardware and software development, often referred to HW/SW co-design. Here, both hardware and software are developed in parallel manner, imposing new design challenges to the system architects.To cope with this issue during MPSoC design we have proposed a higher abstraction level, where an abstract processor simulator called Virtual Processing Unit (VPU) is utilized. | 5 |
| Norwegian University of Science and Technology | Saeeid Tahmasbi Oskuii | Low Power Partial-Product Reduction-Tree Generator for Parallel Multipliers | In this demonstration, a method will be presented for generation of power-efficient parallel multipliers in such a way that its partial products are connected to minimize activity. At each stage of the reduction tree, a simulated annealing optimizer uses power cost numbers from a specially implemented probabilistic gate-level power estimator and selects a power-efficient solution. | 6 |
| University of Tehran | Parisa Razaghi | A Platform for Multi-Language Mixed-Signal Simulation | Mixed-Signal Simulator is a mixed-signal, mixed-domain, and mixed-language design environment which supports VHDL-AMS 1999, VHDL-2002, Verilog 2001, SystemVerilog 2005 assertions, and SystemC 2005. | 7 |
| 14:00 - 16:00 | Session | Physical Design, Modelling and Exploration | UB 3.3 | |
| Université de Bretagne Sud | Pierre Bomel | A mixed (hardware and software) rapid prototyping platform. | PALMYRE is set of hardware components and software tools enabling the rapid prototyping of telecom-oriented application. DSPs from Texas Instruments and FPGAs from Xilinx can be freely mixed to build a specific computing and communication topology. The demonstration is a pipeline of two DSPs and an FPGA Viterbi decoder. Thanks to the GALS/LIS approach the system is data regulated and can be stopped and restarted without loss of any data or synchronization between DSPs and FPGA. | 1 |
| Université de Bretagne Sud | Pierre Bomel | A lightweight and remote partially reconfigurable platform | We present the concept of a networked lightweight and partially reconfigurable platform assisted by a remote bitstreams server. We propose a software and hardware architecture as well as a new data-link level network protocol implementation dedicated to dynamic and partial reconfiguration of FPGAs. These systems target embedded applications with very scarce hardware resources taking advantage of dynamic, specific and optimized architectures. | 2 |
| Universidad Complutense de Madrid | Francisco Vallejos | OS based Wireless Body Area Network for ECG | We present a complete architecture of a Wireless Body Area Network for ECG (electrocardiogram) monitoring.The network is composed of five nodes and a base station. The nodes proccess the ECG signal that they senseand they send information to the base station when a special event occurs, depending on the application. Wehave developed a new TDMA protocol that reduces the energy consumption associated with the radio. | 3 |
| Technical University of Braunschweig | Steffen Stein | Advances in Symbolic Performance Analysis | We will demonstrate the latest advances in formal performance analysis and its application for runtime performance control. We will demonstrate approaches to embedded performance control, current research in event stream modeling and analysis approaches applicable for MPSoC. | 4 |
| University of Tehran | Homa Alemzadeh | SystemC Studio: Translation for TLM Combined Simulation and Synthesis | SystemC Studio provides an interface between RT level VHDL/Verilog descriptions and high level descriptions in SystemC and SystemC TLM. This is done by translating VHDL/Verilog codes into RT level SystemC, linking them with post-synthesis SystemC codes coming from synthesis of high level TLM descriptions, and generating a SystemC/C++ model of the entire hardware to be simulated and verified. This environment also provides SystemC translation to VHDL for RTL synthesis of post high-level synthesis TLM descriptions. | 5 |
