Efficient protection techniques against SEUs for adaptive filters: An echo canceller case study

Abstract

In this paper, novel protection techniques against soft errors for adaptive filters are presented. The new techniques are based on the use of system knowledge in terms of both filter structure and functionality, as well as the application tolerance to soft errors. Adaptive filters by nature recover from soft errors on their coefficients, but in existing implementations the recovery time can exceed what is acceptable for many applications. The proposed techniques dramatically reduce the recovery time after a soft error with an acceptable increment on circuit complexity, as they rely on reusing existing logic. To illustrate these techniques, a case study is presented in which their effectiveness is evaluated using a software-based fault injection platform. Also, their complexity is estimated in terms of the number of equivalent gates generated for the synthesized circuit implementation using a commercial ASIC library.