Secure Wire Shuffling in the Probing Model

In this paper we describe the first improvement of wire shuffling countermeasure against side-channel attacks described by Ishai, Sahai and Wagner at Crypto 2003. More precisely, show how to get worst case statistical security t probes with running time $${\mathcal O}(t)$$ instead O}(t \log t)$$ ; our construction is also much simpler. Recall that classical masking achieves perfect but O}(t^2)$$ . We a practical implementation for AES outperforms $$t \ge 6\,000$$