Hardware-software co-synthesis of high performance embedded computers and development of smart home embedded systems for the elderly

嵌入式计算系统是支持整个系统功能的更大系统的一部分。嵌入式计算机系统的应用范围从复杂的飞机飞行控制到消费产品。传统上，嵌入式计算机已经被开发为两个流的过程，其中硬件工程师提供由软件工程师编程的专用计算机。当硬件和软件彼此适当地“调整”时，可以实现最佳性能。硬件/软件协同设计方法联合收割机将硬件和软件设计工作结合起来，以更短的上市时间生产高效的嵌入式系统。协同综合是硬件/软件协同设计所包含的问题之一，它涉及自动（或半自动）设计硬件和软件以满足系统规范。协同合成过程开始于从任务图到处理元素的任务分配（分配），然后确认系统是否满足要求（通常通过调度）。我们专注于分布式存储器嵌入式计算机，其中一些分布式存储器拓扑结构（如超立方体，二叉和四叉树）提供灵活和可靠的架构。 本研究的动机是需要自动化的嵌入式系统设计过程中，以生产高性能的设备。特别受益于高水平性能和容错的嵌入式应用包括航空航天、医疗仪器和安全关键系统。协同综合技术被用于设计高性能的嵌入式系统，用于医疗设备、智能家居和工业嵌入式应用。我们正在研究基于有源RFID（射频识别）标签的嵌入式系统，用于对象监控和跟踪智能家居的居住者。人们可以将RFID标签贴在老年人的鞋子或衣服上，然后实时监控它们。卫生保健提供者可以使用这样的系统来监测老年人以及痴呆症患者。这将有助于老年人在家里生活更长的时间，有助于提高加拿大人的生活质量。

An embedded computing system is part of a larger system that supports the functioning of overall system. Applications of embedded computer systems range from sophisticated aircraft flight control to consumer products. Traditionally, embedded computer have been developed as a two-stream process, where hardware engineers deliver special-purpose computers that are programmed by software engineers. Optimal performance is achieved when hardware and software are properly 'tuned' to each other. Hardware/Software co-design methodologies combine the hardware and software design efforts to produce efficient embedded systems with lower time to market. The co-synthesis is one of the problems encompassed by hardware/software co-design that deals with automatic (or semi-automatic) design of hardware and software to meet the system specification. Co-synthesis process starts with task assignment from a task graph to processing elements (allocation) followed by confirmation of whether the system meets requirements (usually through scheduling). We are focusing on distributed-memory embedded computers, where some distributed memory topologies (e.g. hypercube, binary and quad-tree) provide flexible and reliable architectures.    This research is motivated by the need to automate the embedded system design process to produce high performance devices. Embedded applications that would particularly benefit from high levels of performance and fault-tolerance include aerospace, medical instrumentation and safety critical systems. The co-synthesis techniques are being employed to design high performance embedded systems for medical devices, smart home, and industrial embedded applications. We are investigating active-RFID (Radio Frequency Identification) tags based embedded systems for object monitoring and tracking the occupants of smart homes. One could attach RFID tags to shoes or clothes of the elderly persons, and then monitor them in real-time. Health care provider can use such systems to monitor elderly as well as people with dementia. It would help the elderly to live at home longer, contributing to a better quality of life for the Canadians.