Abstract

Network Control Systems (NAC) have been used in many industrial processes. They aim to reduce the human factor burden and efficiently handle the complex process and communication of those systems. Supervisory control and data acquisition (SCADA) systems are used in industrial, infrastructure and facility processes (e.g. manufacturing, fabrication, oil and water pipelines, building ventilation, etc.) Like other Internet of Things (IoT) implementations, SCADA systems are vulnerable to cyber-attacks, therefore, a robust anomaly detection is a major requirement. However, having an accurate anomaly detection system is not an easy task, due to the difficulty to differentiate between cyber-attacks and system internal failures (e.g. hardware failures). In this paper, we present a model that detects anomaly events in a water system controlled by SCADA. Six Machine Learning techniques have been used in building and evaluating the model. The model classifies different anomaly events including hardware failures (e.g. sensor failures), sabotage and cyber-attacks (e.g. DoS and Spoofing). Unlike other detection systems, our proposed work focuses on notifying the operator when an anomaly occurs with a probability of the event occurring. This additional information helps in accelerating the mitigation process. The model is trained and tested using a real-world dataset.

Abstract

The Automatic Dependent Surveillance-Broadcast (ADS-B) system is a key component of the Next Generation Air Transportation System (NextGen) that manages the increasingly congested airspace. It provides accurate aircraft localization and efficient air traffic management and also improves the safety of billions of current and future passengers. While the benefits of ADS-B are well known, the lack of basic security measures like encryption and authentication introduces various exploitable security vulnerabilities. One practical threat is the ADS-B spoofing attack that targets the ADS-B ground station, in which the ground-based or aircraft-based attacker manipulates the International Civil Aviation Organization (ICAO) address (a unique identifier for each aircraft) in the ADS-B messages to fake the appearance of non-existent aircraft or masquerade as a trusted aircraft. As a result, this attack can confuse the pilots or the air traffic control personnel and cause dangerous maneuvers. In this paper, we introduce SODA - a two-stage Deep Neural Network (DNN)-based spoofing detector for ADS-B that consists of a message classifier and an aircraft classifier. It allows a ground station to examine each incoming message based on the PHY-layer features (e.g., IQ samples and phases) and flag suspicious messages. Our experimental results show that SODA detects ground-based spoofing attacks with a probability of 99.34%, while having a very small false alarm rate (i.e., 0.43%). It outperforms other machine learning techniques such as XGBoost, Logistic Regression, and Support Vector Machine. It further identifies individual aircraft with an average F-score of 96.68% and an accuracy of 96.66%, with a significant improvement over the state-of-the-art detector. Comment: Accepted to IEEE CNS 2019

Abstract

We present a photonic system that exploits the speckle generated by the interaction of a laser source and a semi-transparent scattering medium, in our case a large-core optical fiber, as a physical root of trust for cryptographic applications, while the same configuration can act as a high-rate machine learning paradigm.

Abstract

The scope of this paper is to demonstrate a fully working and compact photonic Physical Unclonable Function (PUF) device capable of operating in real life scenarios as an authentication mechanism and random number generator. For this purpose, an extensive experimental investigation of a Polymer Optical Fiber (POF) and a diffuser as PUF tokens is performed and the most significant properties are evaluated using the proper mathematical tools. Two different software algorithms, the Random Binary Method (RBM) and Singular Value Decomposition (SVD), were tested for optimized key extraction and error correction codes have been incorporated for enhancing key reproducibility. By taking into consideration the limitations and overall performance derived by the experimental evaluation of the system, the designing details towards the implementation of a miniaturized, energy efficient and low-cost device are extensively discussed. The performance of the final device is thoroughly evaluated, demonstrating a long-term stability of 1 week, an operating temperature range of 50C, an exponentially large pool of unique Challenge-Response Pairs (CRPs), recovery after power failure and capability of generating NIST compliant true random numbers.

Abstract

The linear and grid based Wireless Sensor Networks (WSN) are formed by applications where objects being monitored are either placed in linear or grid based form. E.g. monitoring oil, water or gas pipelines; perimeter surveillance; monitoring traffic level of city streets, goods warehouse monitoring. The security of data is a critical issue for all such applications and as the devices used for the monitoring purpose have several resource constraints (bandwidth, storage capacity, battery life); it is significant to have a lightweight security solution. Therefore, we consider symmetric key based solutions proposed in the literature as asymmetric based solutions require more computation, energy and storage of keys. We analyse the symmetric ciphers with respect to the performance parameters: RAM, ROM consumption and number of CPU cycles. We perform this simulation analysis in Contiki Cooja by considering an example scenario on two different motes namely: Sky and Z1. The aim of this analysis is to come up with the best suited symmetric key based cipher for the linear and grid based WSN. Comment: Cryptography and Information Security (CRIS-2018)

Abstract

We present a novel approach to mitigate buffer overflow attack using Variable Record Table (VRT). Dedicated memory space is used to automatically record base and bound information of variables extracted during runtime. We instrument frame pointer and function(s) related registers to decode variable memory space in stack and heap. We have modified Simplescalar/PISA simulator to extract variables space of six (6) benchmark suites from MiBench. We have tested 290 small C programs (MIT corpus suite) having 22 different buffer overflow vulnerabilities in stack and heap. Experimental results show that our approach can detect buffer overflow attack with zero instruction overhead with the memory space requirement up to 13Kb to maintain VRT for a program with 324 variables. Comment: Accepted for publication in MWSCAS2019

Abstract

Searching for available parking spaces is a major problem for drivers in big cities, causing traffic congestion and air pollution, and wasting drivers' time. Smart parking systems enable drivers to have real-time parking information for pre-booking. However, current smart parking requires drivers to disclose their private information, such as desired destinations. Moreover, the existing schemes are centralized and vulnerable to the bottleneck of the single point of failure and data breaches. In this paper, we propose a distributed privacy-preserving smart parking system using blockchain. A consortium blockchain created by different parking lot owners to ensure security, transparency, and availability is proposed to store their parking offers on the blockchain. To preserve drivers' location privacy, we adopt private information retrieval (PIR) technique to enable drivers to retrieve parking offers from blockchain nodes privately, without revealing which parking offers are retrieved. Furthermore, a short randomizable signature is used to enable drivers to reserve available parking slots in an anonymous manner. Besides, we introduce an anonymous payment system that cannot link drivers' to specific parking locations. Our performance evaluations demonstrate that the proposed scheme can preserve drivers' privacy with low communication and computation overhead.

Abstract

Network function virtualization (NFV) based service function chaining (SFC) allows the provisioning of various security and traffic engineering applications in a cloud network. Inefficient deployment of network functions can lead to security violations and performance overhead. In an OpenFlow enabled cloud, the key problem with current mechanisms is that several packet field match and flow rule action sets associated with the network functions are non-overlapping and can be parallelized for performance enhancement. We introduce Network Function Parallelism (NFP) SFC-NFP for OpenFlow network. Our solution utilizes network function parallelism over the OpenFlow rules to improve SFC performance in the cloud network. We have utilized the DPDK platform with an OpenFlow switch (OVS) for experimental analysis. Our solution achieves a 1.40-1.90x reduction in latency for SFC in an OpenStack cloud network managed by the SDN framework. Comment: 5 page