Paul Kocher recently developped attacks based on the electric consumption of chips that perform cryptographic computations. Among those attacks, the "Differential Power Analysis" (DPA) is probably one of the most impressive and most difficult to avoid. In this paper, we present several ideas to resist this type of attack, and in particular we develop one of them which leads, interestingly, to rather precise mathematical analysis. Thus we show that it is possible to build an implementation that is provably DPA-resistant, in a "local" and restricted way (i.e. when-given a chip with a fixed key-the attacker only tries to detect predictable local deviations in the differentials of mean curves). We also briefly discuss some more general attacks, that are sometimes efficient whereas the "original" DPA fails. Many measures of consumption have been done on real chips to test the ideas presented in this paper, and some of the obtained curves are printed here.

Abstract. In [16], J. Patarin designed a new scheme, called "Oil and Vinegar", for computing asymmetric signatures. It is very simple, can be computed very fast (both in secret and public key) and requires very little RAM in smartcard implementations. The idea consists in hiding quadratic equations in n unknowns called "oil" and v = n unknowns called "vinegar" over a finite field K, with linear secret functions. This original scheme was broken in [10] by A. Kipnis and A. Shamir. In this paper, we study some very simple variations of the original scheme where v > n (instead of v = n). These schemes are called "Unbalanced Oil and Vinegar" (UOV), since we have more "vinegar" unknowns than "oil" unknowns. We show that, when v n, the attack of [10] can be extended, but when v ≥ 2n for example, the security of the scheme is still an open problem. Moreover, when v n 2 2 , the security of the scheme is exactly equivalent (if we accept a very natural but not proved property) to the problem of solving a random set of n quadratic equations in n 2 2 unknowns (with no trapdoor). However, we show that (in characteristic 2) when v ≥ n 2 , finding a solution is generally easy. Then we will see that it is very easy to combine the Oil and Vinegar idea and the HFE schemes of [14]. The resulting scheme, called HFEV, looks at the present also very interesting both from a practical and theoretical point of view.

Sosemanuk is a new synchronous software-oriented stream cipher, corresponding to Profile 1 of the ECRYPT call for stream cipher primitives. Its key length is variable between 128 and 256 bits. It accommodates a 128-bit initial value. Any key length is claimed to achieve 128-bit security. The Sosemanuk cipher uses both some basic design principles from the stream cipher SNOW 2.0 and some transformations derived from the block cipher SERPENT. Sosemanuk aims at improving SNOW 2.0 both from the security and from the efficiency points of view. Most notably, it uses a faster IV-setup procedure. It also requires a reduced amount of static data, yielding better performance on several architectures.

No abstract

As Elliptic Curve Cryptosystems are becoming more and more popular and are included in many standards, an increasing demand has appeared for secure implementations that are not vulnerable to sidechannel attacks. To achieve this goal, several generic countermeasures against Power Analysis have been proposed in recent years. In particular, to protect the basic scalar multiplication-on an elliptic curve-against Differential Power Analysis (DPA), it has often been recommended using "random projective coordinates", "random elliptic curve isomorphisms" or "random field isomorphisms". So far, these countermeasures have been considered by many authors as a cheap and secure way of avoiding the DPA attacks on the "scalar multiplication" primitive. However we show in the present paper that, for many elliptic curves, such a DPA-protection of the "scalar" multiplication is not sufficient. In a chosen message scenario, a Power Analysis attack is still possible even if one of the three aforementioned countermeasures is used. We expose a new Power Analysis strategy that can be successful for a large class of elliptic curves, including most of the sample curves recommended by standard bodies such as ANSI, IEEE, ISO, NIST, SECG or WTLS. This result means that the problem of randomizing the basepoint may be more difficult than expected and that "standard" techniques have still to be improved, which may also have an impact on the performances of the implementations.