An overview is presented of the main objectives and current achievements of the architecture synthesis work within part of the Basic Research Action ASCIS (architecture synthesis for complex integrated systems) project. The main goal of this work is to contribute to the solution of one of the major bottlenecks restricting the use of ASICs in industrial systems, namely, the lack of efficient design methodologies and corresponding CAD techniques which support the development of cost-effective application-specific architectures for given throughput or latency. The following areas of the project are emphasized: novel design methodologies and CAD techniques for partitioning, memory management, allocation, binding and scheduling; generic global optimization techniques to support those synthesis approaches; and performance-driven controller synthesis.< >
The VLSI image processor, IPRISC, designed specifically for real-time feedback control of robots involved in small-part assembly in a flexible manufacturing system, is presented. A brief overview of robots in flexible manufacturing systems is given, and the need for an intelligent feedback mechanism for the robot is stressed. It is asserted that vision provides the most accurate and flexible form of feedback but has the disadvantage of requiring enormous amounts of data throughout. For the robot to operate in real time requires an architecture designed to execute optimally the image-processing algorithms relevant to inline robotic control.< >
In this paper we show that by performing some VHDL elaboration transformations before synthesis we can extend the synthesis subset to include complex structural and hierarchical statements. This in turn means that: design, debug and simulation times are reduced; designs are more accessible (readable, modifiable, portable, reusable); design prototyping can be speeded up. All of this can be achieved without the need to modify existing synthesis tools.
This article explores the impact of online Irish traditional singing sessions on health and well-being during the COVID-19 pandemic. Singing sessions are unique facets of Ireland's music tradition that saw dramatic closure, interruption and digital transition in response to COVID-19 social distancing measures. This study highlights a gap in health promotion literature with regard to traditional singing sessions as a group singing activity and examines the potential for online group singing activities to have positive impacts on the health and well-being of participants. While traditional singing sessions foreground solo performances, they are quintessentially group activities, and include community engagement and active participation from singers and listeners alike. Through an online survey (n = 108), and ethnographic interviews (n = 3), this study explores potential health and well-being implications of online traditional singing sessions, and reveals four main areas of impact: social connection, enjoyment, cognitive motivation and timekeeping. The study suggests that online traditional singing sessions can promote health and well-being in participants, particularly during times of isolation.This article explores the impact of online Irish traditional singing sessions on health and well-being during the COVID-19 pandemic. Singing sessions are unique facets of Ireland’s music tradition which were forced to move online due to COVID-19 restrictions. This study used an online survey (n = 108), and interviews (n = 3), to explore the impact of these online sessions on the well-being of their participants. Findings showed the impacts to be overwhelmingly positive, particularly in four main areas: social connection, enjoyment, cognitive motivation, and timekeeping. This study highlights the value of traditional singing sessions as group singing activities for the purposes of health and well-being promotion, and suggests that online group singing activities can be beneficial, particularly during times of isolation.
This paper presents a method for modeling and synthesizing mixed HW/SW systems. The proposed method starts from a full system-level specification. Systems are modeled in a synthesis-oriented manner by means of an extended finite state machine model. System-level synthesis is composed of three tasks: partitioning systems into inter-dependent sub-systems, inter-subsystem communication synthesis and architecture mapping onto a flexible architecture platform which includes both hardware and software components. The overall method is illustrated with an example.< >
In this paper, we present a methodology whereby the whole synthesis and prototyping cycle can be speeded up simply by extending the acceptable VHDL subset to include hitherto unsynthesisable constructs. VHDL elaboration transformations as well as some compiler optimisation techniques can be performed to ensure that the VHDL model is still acceptable by commercial synthesis tools. The advantages of this methodology are shown using a real industrial application: the development of a generic VHDL memory model for fast system reconfiguration in a hardware emulation environment.
