Leakage-Resilience Meets Incompressibility

Abstract

Leakage-resilient cryptography aims to maintain security even if an adversary gains partial knowledge of the secret key while having access to the entire ciphertext. In typical scenarios, this involves designing systems that can withstand side-channel attacks, where small but potentially crucial bits of the secret key might be leaked through unintended channels such as timing information or power consumption patterns. Incompressible cryptography, on the other hand, offers security in situations where the adversary can obtain the entire secret key but only a portion of the ciphertext. This type of cryptography is useful in settings where the adversary’s ability to capture and store data is limited. For example, in scenarios where large data streams or files are encrypted, and the adversary can only intercept fragments.

However, cryptographic schemes that are secure in one of these models may not necessarily be secure in the other. In this talk, we introduce a new security model called leakage-resilient incompressible cryptography that integrates both leakage-resilient and incompressible cryptography. By considering both types of threats simultaneously, we can develop encryption schemes that are resilient to a broader range of attacks. We will also discuss advanced forms of incompressible encryption, such as incompressible functional encryption.

Venue: A3 117, Vindhya

Speaker Bio

Mahesh Sreekumar Rajasree is a Post-Doctoral Fellow at IIT Delhi, supervised by Prof. Venkata Koppula. His primary research interests lie in the fields of algorithms and cryptography, with a current focus on public key cryptography. He completed his PhD and MTech from IIT Kanpur under the guidance of Prof. Manindra Agrawal. His PhD thesis primarily involved cryptanalysis of symmetric key cryptosystems and algorithms related to variants of the subset-sum problem, while his MTech thesis contributed new results in lattice theory. He obtained his BTech degree from NIT Calicut.