Automatic dependent surveillance–broadcasting (ADS-B) is a surveillance technology strongly promoted by the International Civil Aviation Organization. It has been widely used in commercial and general aviation, to provide support for the normal operation of air traffic control (ATC) in commercial aviation. However, the openness of the ADS-B protocol makes it extremely vulnerable to cyber attacks. Previous research did not specifically consider the application scenarios of ATC in commercial air transport, and there is a problem of low attack detection rates. This article focuses on ADS-B attack detection under the background of ATC. We combine the flight plan with ADS-B information to construct an airspace flight image stream and process the image stream using a generative adversarial network–long short-term memory (GAN-LSTM) model to predict future images. Then, we identify abnormal images based on the normalized cross-correlation and mark anomalous targets. Our method can quickly locate anomalous targets in the controlled airspace. Based on real flight data, abnormal data for various malicious attacks were forged. We evaluated the detection performance of the method through a confusion matrix and several performance indices. The final experimental results showed that our detection scheme exhibited good detection performances for various attacks, with an average detection accuracy of 92.3%, a false positive rate of 11.9%, and a false negative rate of 6.2%. This approach guarantees the information security of ADS-B, thereby improving the operational security of ATC.
12 January 2026 in IEEE Transactions on Aerospace and Electronic Systems
As a widely deployed aeronautical surveillance technology, Automatic Dependent Surveillance-Broadcast (ADS-B) has become vulnerable to various security threats due to its reliance on plaintext data transmission through open protocols and unencrypted communication channels. Existing security schemes for ADS-B systems face significant challenges, including high communication overhead and excessive computational delay. To address these limitations and achieve efficient lightweight authentication for ADS-B data, this study proposes a novel intra-inter chain framework. Building upon this framework, a tree-chain authentication scheme using SM3 and SM9 cryptographic algorithms is desiged. This scheme organizes ADS-B data packets into authentication chains, employs a tree hash model for hash generation, and constructs a shared signature for multiple chains. Comprehensive theoretical analysis and extensive experiments are conducted to validate the scheme's functionality and evaluate its performance. Experimental results show that our authentication scheme can effectively defend against ADS-B data spoofing attacks. Compared with other existing authentication schemes, the proposed scheme reduces communication overhead by approximately 19.3% and decreases hash/signature generation time and data verification delays by at least 42.8% and 16.5%, respectively. Moreover, the scheme maintains a verification probability of 98.7% even under high packet loss conditions, exhibiting exceptional robustness. This study provides a practical and efficient solution to enhance the security of ADS-B surveillance systems.
Software-Defined Networking (SDN) actualizes the separation of control and forwarding and innovates network functionalities with a logically centralized controller. Contemporary SDN infrastructure exposes the potential bottlenecks which are prone to engage in distributed denial of service attack (DDoS) thus posing an ever-increasing threat. This paper adopts the idea of “cross-plane collaboration” accomplishing DDoS attack defense and incorporates a two-phase project deploying the lightweight detection mechanism in data layer and the fine-grained filtering model in control layer. The coadjutant detection mechanism introduces a novel three-dimensional entropy consisting of five flow table features performing real-time feature detection; the defense strategy schedules an attack classification algorithm based on neural network by means of extracting four flow rule features designed to locate compromised interfaces occupied by malicious traffic. Extensive experiments are implemented to demonstrate the method we proposed brings excellent superiority. The detection rate of the classification filtering model is 99.4%, and it is real-time, with a detection time of 0.51s. In addition, the method of cross-layer defense reduces the CPU utilization of the controller.