Towards robust and low-complexity radiometric fingerprint

Authentication is challenging in the IoT because most advanced cryptographic algorithms are difficult to afford by constrained devices expected to run many months or years. Radiometric signatures have been used effectively to identify wireless devices based on imperfections in electronic circuits. This technique is also known as Radio frequency (RF) fingerprinting. Previous work proves the feasibility of this technique but mainly considered static channel conditions. In our current work, we systematically and experimentally study the impact of dynamic and complex channel conditions on the radiometric signatures. The results show a threat to identification accuracy for modulation error-based fingerprinting that was considered channel-resilient. Next step, my research aims at improving the robustness of RF fingerprint systems and make this authentication solution ready for widespread implementation.