Low-Cost Hardening of Image Processing Applications Against Soft Errors

Image processing systems are increasingly used in safety-critical applications, and their hardening against soft errors becomes an issue. The authors propose a methodology to identify soft errors as uncritical based on their impact on the system's functionality. The authors call a soft error uncritical if its impact is provably limited to image perturbations during a very short period of time (number of cycles) and the system is guaranteed to recover thereafter. Uncritical errors do not require hardening as their effects are unperceivable for the human user of the system. The authors focus on soft errors in the motion estimation subsystem of MPEG-2 and introduce different definitions of uncritical soft errors in that subsystem. A method is proposed to automatically determine uncritical errors and provide experimental results for various parameters. The concept can be adapted to further systems and enhance existing methods