Intrusion Detection for ARINC-429: A Hybrid FFT and Unsupervised Learning Approach

ARINC-429 is a widely-used avionics communication standard in both commercial and military aircraft. Originally designed with a strong emphasis on safety, ARINC-429 did not incorporate security mechanisms such as authentication or encryption, leaving it vulnerable to cybersecurity threats in increasingly interconnected avionics systems. This study introduces a hybrid approach to enhance ARINC-429 security through hardware fingerprinting. The proposed method combines the Fast Fourier Transform as an initial layer for anomaly filtering with the Local Outlier Factor as a second layer to reduce false positives and make final anomaly detection decisions. Experimental results demonstrate that this layered approach effectively balances computational efficiency with high detection accuracy, making it suitable for real-time intrusion detection in avionics environments.