Cryptographic modules in wireless sensor network (WSN) nodes often face side channel attack threat. However, since the data encrypted in WSN nodes comes from natural physical signals, the distribution of plaintext will follow their natural properties, in all probability instead of being uniformly random. In order to explore the success rate of differential power analysis (DPA) when applied to nonuniform-distributed data from real sensors, we theoretically and experimentally analyze the relationship between the distribution of plaintext and the success rate of DPA on block ciphers. It turns out that the success rate of DPA targeting the first round is negatively correlated to the Cramer-von-Mises distance between the plaintext's distribution and the uniform distribution. The conclusion infers that if the attacker can construct plaintext, the uniform data makes the biggest success rate, and if the attacker only has access to real data, attacking the last round is better than attacking the first round.

 

side channel analysis, WSN, DPA, success rate, distribution of plaintext