An ADS-B spoofing attack detection method based on LASSO ensemble empirical mode decomposition

As the heart of modern air traffic control, automatic dependent surveillance-broadcast (ADS-B) features less aircraft separation required and lower cost when comparing with secondary surveillance radar(SSR). Different from advocacy-response mode for SSR, ADS-B is constraint with its security since anyone with the off-the-shelf communication equipment can transmit their own false data. In this work, a novel method based on signal processing method was proposed to detect cyber attack which was widely concerned with the deadline for ADS-B implementation in most airspace approaching. The LASSO regression framework was proposed to break the constraint of uniform weight in the conventional ensemble empirical mode decomposition(EEMD). Based on the proposed method, the ADS-B signal was then implemented to LASSO EEMD. The obtained IMFs were analyzed to determine the feature that distinguishes the real ADS-B with the artificial interference. A simulation system composing of ADS-B receiver and spoofing sources produced by vehicle ADS-B out was build. The results show that the proposed method was efficient by a series of artificially spoofing attacks.