According to recent security analysis reports, malicious software (a.k.a. malware) is rising at an alarming rate in numbers, complexity, and harmful purposes to compromise the security of modern computer systems. Recently, malware detection based on low-level hardware features (e.g., Hardware Performance Counters (HPCs) information) has emerged as an effective alternative solution to address the complexity and performance overheads of traditional software-based detection methods. Hardware-assisted Malware Detection (HMD) techniques depend on standard Machine Learning (ML) classifiers to detect signatures of malicious applications by monitoring built-in HPC registers during execution at run-time. Prior HMD methods though effective have limited their study on detecting malicious applications that are spawned as a separate thread during application execution, hence detecting stealthy malware patterns at run-time remains a critical challenge. Stealthy malware refers to harmful cyber attacks in which malicious code is hidden within benign applications and remains undetected by traditional malware detection approaches. In this paper, we first present a comprehensive review of recent advances in hardware-assisted malware detection studies that have used standard ML techniques to detect the malware signatures. Next, to address the challenge of stealthy malware detection at the processor’s hardware level, we propose StealthMiner, a novel specialized time series machine learning-based approach to accurately detect stealthy malware trace at run-time using branch instructions, the most prominent HPC feature. StealthMiner is based on a lightweight time series Fully Convolutional Neural Network (FCN) model that automatically identifies potentially contaminated samples in HPC-based time series data and utilizes them to accurately recognize the trace of stealthy malware. Our analysis demonstrates that using state-of-the-art ML-based malware detection methods is not effective in detecting stealthy malware samples since the captured HPC data not only represents malware but also carries benign applications’ microarchitectural data. The experimental results demonstrate that with the aid of our novel intelligent approach, stealthy malware can be detected at run-time with 94% detection performance on average with only one HPC feature, outperforming the detection performance of state-of-the-art HMD and general time series classification methods by up to 42% and 36%, respectively.
4 October 2018 in IEEE Transactions on Intelligent Vehicles
Automatic Dependent Surveillance Broadcast (ADS- B) is one the most prominent air traffic surveillance technologies that emerged in recent years. It can be used not only as a complement to existing radar-based technologies to locate air- craft, but also as a replacement in areas without radar coverage. However, a major flaw in the deployment of this technology is its reliance on clear-text, open broadcast of messages, which makes it vulnerable to attackers. In this paper, we propose ADS-Bsec, a holistic framework that solves the root cause of ADS-B problems by enforcing the much needed authenticity and integrity in ADS- B messages. It does so without changing the message format, and along with a proper key exchange mechanism. In addition, our framework also leverages results from radio-location and risk- adaptive techniques in order to prevent attacks targeting ADS- B. An important characteristic of our framework is that it was mostly implemented using a test bed with actual prototype radios, which mimics as much as possible real flight conditions. As a result, our framework is a promising approach for integrating ADS-B to NextGen infrastructure in a secure and affordable fashion.
Recent electronics advances for air transport have increased aircraft density, volume, and frequency in the airspace. These advances come with control requirements for precise navigation, coordinated Air Traffic Management (ATM) or Unmanned aircraft system Traffic Management (UTM), and proactive security. The tight tolerances of aircraft control necessitate management of spatial uncertainty, timeliness precision, and confidence assessment, which have, respectively, variance, reliability, and veracity situation awareness and assessment metrics. Meeting such airspace requirements involves the ability to evaluate how those metrics impact ATM/UTM operations, making the complex interrelationships between them a key aspect for coping with the fast worldwide growth of air transport. To support such growth, ontologies have been proposed as a promising technology for making such interrelationships explicit, while facilitating communication between avionics devices. This paper investigates the use of ontologies in support of electronic ATM/UTM operations, highlighting the use of Uncertainty Representation and the Reasoning Evaluation Framework (URREF) in realizing the ability for Air Traffic Controllers (ATCs) to semantically communicate with aircraft operators concerning physical airspace coordination. Using Avionics Analytics Ontology (AAO) endowed with the URREF, application examples based on two airspace situations are presented. Example results for northeast coast of Brazil atmospheric volcanic ash as well as for the Eyjafjallajokull volcano eruption show a 65–80% success in providing warnings to ATCs for airspace control. The paper demonstrates that an ontology-based UTM enhances the capability and accuracy of an ATM to suggest rerouting in the presence of remarkably deteriorated weather conditions.
The Federal Aviation Administration (FAA) and the European Aviation Safety Agency (EASA) have an emerging interest in ontologies. A common ontology for avionics analytics can help pilots and Air Traffic Controllers (ATCs) make difficult decisions with increasingly-sophisticated avionics, densely-occupied airspaces, and progressively-adverse weather. The connected airspace sets a context of big data, enriched features, and information uncertainty. This paper proposes to endow an Avionics Analytics Ontology (AAO) with semantic uncertainty to improve decision-making capabilities. The proposed approach incorporates the Uncertainty Representation and Reasoning Evaluation Framework (URREF) into the AAO for input information. The AAO focuses on veracity as a key component of information credibility to deal with uncertainty. The URREF assessment aims to enhance avionics analytics when considering semantic and physical data sources. Thus, situation AWareness (SAW) and Situation Assessment (SA) as well as Situation Understanding (SU) in information fusion are ultimately enhanced by means of statistical metrics of veracity. This paper also shows experimental results from two application scenarios. Concluding remarks and future research directions are also presented.
10 July 2025 in IEEE Transactions on Intelligent Transportation Systems
In recent years, the use of Unmanned Aircraft Systems (UAS), also known as drones, has been on the rise covering several areas that span from surveillance to retail delivery of goods. This proliferation can be accredited to various advances such as low-latency connectivity, fast execution of control commands, high reliability and more importantly their potential for autonomous operations. Consequently, this allowed the information fusion community to develop various methods of fusion of different types of data. However, these methods in general do not thoroughly consider the proliferation of cyber attacks targeting UAS vehicles especially for civilian and military missions. In this paper, we propose a novel approach leveraging information fusion to secure autonomous drone operations entitled Information Fusion for Secure Drone Operations (IFSDO). Our solution allows the distribution and enforcement of security policies and dynamic spectrum allocation policies to drones during all mission phases depending on the user role. IFSDO works by orchestrating different techniques including cryptography, attribute-based access control and ontological rules to allow the drone to autonomously and securely manage various types of access requests to ensure the mission can be conducted safely. We validate our findings with a case study illustrating the initial exchange of critical data between the security manager and the drone as part of a civilian package delivery mission involving several stakeholders interacting with the drones. The initial results show promising potential of success of our approach.