Robustness with respect to SEU of a hardware-implemented self-converging algorithm

Abstract

Self-convergence is a property that allows distributed systems, when perturbed or badly initialized, to recover a correct operation within finite number of calculation steps. In this paper, an FPGA implementation of this algorithm is presented. The intrinsic robustness of the studied implementation with respect to soft errors resulting from radiation effects is explored by means of a fault-injected method. Obtained results put in evidence the fault-tolerance capabilities and robustness of the tested hardware-implemented algorithm.