Emerging Resilience Techniques For Embedded Devices

This chapter presents a broad range of techniques for highly reliable and survivable field programmable gate array-based embedded computing systems operating in harsh environments. The notion of autonomous self-repair is essential for such systems as physical access to such platforms is often limited. In this regard, adaptable reconfiguration-based techniques are presented in this chapter, whereby an appropriate set of resources can be selected at runtime to achieve graceful performance degradation in the presence of multiple permanent and transient faults. Intermittent quality of repair (QoR) can be achieved by use of proposed genetic algorithm-based intelligent techniques for relatively compact circuits as compared to conventional approaches. Video encoder use cases are used as benchmarks, and significance driven resource allocation is utilized to demonstrate benefit in terms of recovery time, throughput degradation, and QoR. These metrics can be utilized as a gauge for evaluating the efficacy of multiple FH techniques. Finally, a discussion of real-time tradeoffs of resilience, quality, and energy for proposed approaches is presented