Perancangan dan Implementasi Sistem Pembangkitan Kunci Kriptografi Deterministik Berbasis Biometrik Sidik Jari pada Android

Abstract

Digital data security on Android platforms relies heavily on cryptography, yet conventional authentication methods such as passwords are increasingly vulnerable to cyberattacks and prone to user forgetfulness. Fingerprint biometric authentication offers a secure alternative, but it faces inherent constraints due to the fuzzy nature of biological data, which is inconsistent with the rigid requirements of deterministic cryptographic keys. Furthermore, modern smartphone sensors have physical limitations, capturing only 10 to 15 percent of the fingerprint area partially. This research aims to design and build a mobile application using the Flutter framework to bridge these biometric reading inconsistencies. The development process utilizes the Waterfall model, including analysis, design, implementation, and testing phases. The system delegates biometric validation to the Android native Trusted Execution Environment (TEE) to ensure data privacy. Upon successful authentication, the system retrieves Helper Data from a cloud database (Supabase) and concatenates it with the user's identity. This combination is then fused using the SHA-256 one-way hash algorithm to generate a private key. Black-Box testing results demonstrate that the system successfully generates a 256-bit private cryptographic key that is 100% consistent for every scan validated by the TEE. This architectural approach effectively overcomes hardware sensor limitations without exposing raw biological data, making it highly reliable as a high-level passwordless security mechanism on mobile devices.