Reliability analysis of self-repairing bio-inspired cellular hardware

The systematic study of artificial cellular systems, like cellular automata, neural networks or processor arrays, has gained momentum during the past few years in artificial life studies. The goal is to understand the emergent behaviours observed in natural cellular systems. One such behaviour is the ability to tolerate faults. For example, the human body is one of the most complex systems ever known; failures are not rare, but the overall function is highly reliable because of self-diagnosis and self-healing mechanisms that work ceaselessly throughout the body. To borrow the main principles sustaining these mechanisms and applying them to the design of electronic systems could result in a new approach for the design of fault-tolerant systems. Embryonics introduces a new family of fault-tolerant field programmable gate arrays (FPGAs) inspired by nature. Its main ideas come from the mechanisms sustaining the embryonic development of multicellular organisms in nature. Embryonic arrays have also been studied as a technique for improving the fault tolerance properties of cellular architectures. The paper presents mathematical models to assess up to what extent the embryonics architecture improves the reliability of cellular systems