Luca Sterpone - Böcker

What is exactly “Safety”? A safety system should be defined as a system that will not endanger human life or the environment. A safety-critical system requires utmost care in their specification and design in order to avoid possible errors in their implementation that should result in unexpected system’s behavior during his operating “life”. An inappropriate method could lead to loss of life, and will almost certainly result in financial penalties in the long run, whether because of loss of business or because the imposition of fines. Risks of this kind are usually managed with the methods and tools of the “safety engineering”. A life-critical system is designed to 9 lose less than one life per billion (10 ). Nowadays, computers are used at least an order of magnitude more in safety-critical applications compared to two decades ago. Increasingly electronic devices are being used in applications where their correct operation is vital to ensure the safety of the human life and the environment. These application ranging from the anti-lock braking systems (ABS) in automobiles, to the fly-by-wire aircrafts, to biomedical supports to the human care. Therefore, it is vital that electronic designers be aware of the safety implications of the systems they develop. State of the art electronic systems are increasingly adopting progr- mable devices for electronic applications on earthling system. In particular, the Field Programmable Gate Array (FPGA) devices are becoming very interesting due to their characteristics in terms of performance, dimensions and cost.

Embedded systems applications that are either mission or safety-critical usually entail low- to mid- production volumes, require the rapid development of specific tasks, which are typically computing intensive, and are cost bounded. The adoption of re-configurable FPGAs in such application domains is constrained to the availability of suitable techniques to guarantee the dependability requirements entailed by critical applications. This book describes the challenges faced by designers when implementing a mission- or safety-critical application using re-configurable FPGAs and it details various techniques to overcome these challenges. In addition to an overview of the key concepts of re-configurable FPGAs, it provides a theoretical description of the failure modes that can cause incorrect operation of re-configurable FPGA-based electronic systems. It also outlines analysis techniques that can be used to forecast such failures and covers the theory behind solutions to mitigate fault effects. This book also reviews current technologies available for building re-configurable FPGAs, specifically SRAM-based technology and Flash-based technology. For each technology introduced, theoretical concepts presented are applied to real cases. Design techniques and tools are presented to develop critical applications using commercial, off-the-shelf devices, such as Xilinx Virtex FPGAs, and Actel ProASIC FPGAs. Alternative techniques based on radiation hardened FPGAs, such as Xilinx SIRF and Atmel ATF280 are also presented. This publication is an invaluable reference for anyone interested in understanding the technologies of re-configurable FPGAs, as well as designers developing critical applications based on these technologies.

Embedded systems applications that are either mission or safety-critical usually entail low- to mid- production volumes, require the rapid development of specific tasks, which are typically computing intensive, and are cost bounded. The adoption of re-configurable FPGAs in such application domains is constrained to the availability of suitable techniques to guarantee the dependability requirements entailed by critical applications. This book describes the challenges faced by designers when implementing a mission- or safety-critical application using re-configurable FPGAs and it details various techniques to overcome these challenges. In addition to an overview of the key concepts of re-configurable FPGAs, it provides a theoretical description of the failure modes that can cause incorrect operation of re-configurable FPGA-based electronic systems. It also outlines analysis techniques that can be used to forecast such failures and covers the theory behind solutions to mitigate fault effects. This book also reviews current technologies available for building re-configurable FPGAs, specifically SRAM-based technology and Flash-based technology. For each technology introduced, theoretical concepts presented are applied to real cases. Design techniques and tools are presented to develop critical applications using commercial, off-the-shelf devices, such as Xilinx Virtex FPGAs, and Actel ProASIC FPGAs. Alternative techniques based on radiation hardened FPGAs, such as Xilinx SIRF and Atmel ATF280 are also presented. This publication is an invaluable reference for anyone interested in understanding the technologies of re-configurable FPGAs, as well as designers developing critical applications based on these technologies.