Leakage resilient IBE and IPE under the DLIN assumption

Abstract

In this paper, we show identity-based encryption (IBE) and inner product encryption (IPE) schemes which achieve the maximum-possible leakage rate 1-o(1). These schemes are secure under the decision linear (DLIN) assumption in the standard model. Specifically, even if 1-o(1) fraction of each private key is arbitrarily leaked, the IBE scheme is fully secure and the IPE scheme is selectively secure. Mentioned results are in the bounded memory leakage model (Akavia et al., TCC '09). We show that they naturally extends to the continual memory leakage model (Brakerski et al., Dodis et al., FOCS '10). In this stronger model, the leakage rate becomes 1/2-o(1). © 2013 Springer-Verlag.