Fresh Re-keying: Security against Side-Channel and Fault Attacks for Low-Cost Devices

Abstract

The market for RFID technology has grown rapidly over the past few years. Going along with the proliferation of RFID technology is an increasing demand for secure and privacy-preserving applications. In this context, RFID tags need to be protected against physical attacks such as Differential Power Analysis (DPA) and fault attacks. The main obstacles towards secure RFID are the extreme constraints of passive tags in terms of power consumption and silicon area, which makes the integration of countermeasures against physical attacks even more difficult than for other types of embedded systems. In this paper we propose a fresh re-keying scheme that is especially suited for challenge-response protocols such as used to authenticate tags. We evaluate the resistance of our scheme against fault and side-channel analysis, and introduce a simple architecture for VLSI implementation on RFID tags. In addition, we estimate the cost of our scheme in terms of area and execution time for various sec... Abstract. The market for RFID technology has grown rapidly over the past few years. Going along with the proliferation of RFID technology is an increasing demand for secure and privacy-preserving applications. In this context, RFID tags need to be protected against physical attacks such as Differential Power Analysis (DPA) and fault attacks. The main obstacles towards secure RFID are the extreme constraints of passive tags in terms of power consumption and silicon area, which makes the integration of countermeasures against physical attacks even more difficult than for other types of embedded systems. In this paper we propose a fresh re-keying scheme that is especially suited for challenge-response protocols such as used to authenticate tags. We evaluate the resistance of our scheme against fault and side-channel analysis, and introduce a simple architecture for VLSI implementation on RFID tags. In addition, we estimate the cost of our scheme in terms of area and execution time for various security/performance trade-offs. Our experimental results show that the proposed re-keying scheme provides better security (and does so at less cost) than other state-of-the-art countermeasures.

MEAS: memory encryption and authentication secure against side-channel attacks

“…We approach the topic with a detailed analysis of the security of fresh re-keying [33,39] as a promising mechanism to prevent DPA on memory encryption. While re-keying completely thwarts DPA on the cryptographic key, our major result here is that re-keying provides merely first-order DPA security for the memory content itself.…”

“…Therefore, frequent rekeying [33,39] tries to limit the number of different processed inputs per key, i.e., the data complexity.…”

“…1 prevent DPA, but require the initialization with a fresh key and thus secure key synchronization between the communicating parties. A common approach to this synchronization is to derive a fresh key from a shared master secret k and a public, random nonce n [21,39,53]. However, this approach shifts the DPA problem to the key derivation, which thus needs DPA protection through mechanisms like masking.…”

Fresh Re-keying II: Securing Multiple Parties against Side-Channel and Fault Attacks

“…Whereas the scheme is efficient for long messages, the initialization effort might render it impractical for smartcard and RFID applications. More recently, a fresh rekeying scheme was presented at Africacrypt 2010 [19]. It combines the re-keying used in leakage-resilient cryptography with easy to protect low-cost mappings in order to remove the initialization overhead.…”

“…It essentially encrypts every message block m under a fresh session key k * , with a block cipher. The session key is generated from the master key k and a public random nonce r, with a function g. At first sight, it may seem that one just shifts the problem of protecting the block cipher f against physical attacks to the one of protecting the function g. Interestingly, it is argued in [19] that a proper selection of g may lead this scheme to be significantly easier to protect against such attacks than the underlying block cipher f . Namely, g does not need to be cryptographically strong.…”