ABSTRACT
From smart homes to industrial environments, the IoT is an ally to easing daily activities, where some of them are critical. More and more devices are connected to and through the Internet, which, given the large amount of different manufacturers, may lead to a lack of security standards. Denial of service attacks (DDoS, DoS) represent the most common and critical attack against and from these networks, and in the third quarter of 2021, there was an increase of 31% (compared to the same period of 2020) in the total number of advanced DDoS targeted attacks. This work uses the Bot-IoT dataset, addressing its class imbalance problem, to build a novel Intrusion Detection System based on Machine Learning and Deep Learning models. In order to evaluate how the records timestamps affect the predictions, we used three different feature sets for binary and multiclass classifications; this helped us avoid feature dependencies, as produced by the Argus flow data generator, whilst achieving an average accuracy >99%. Then, we conducted comprehensive experimentation, including time performance evaluation, matching and exceeding the results of the current state-of-the-art for identifying denial of service attacks, where the Decision Tree and Multi-layer Perceptron models were the best performing methods to identify DDoS and DoS attacks over IoT networks.
Subject(s)
Deep Learning , Internet of Things , Internet , Machine Learning , Neural Networks, ComputerABSTRACT
The Industrial Internet of Things (IIoT) is considered a key enabler for Industry 4.0. Modern wireless industrial protocols such as the IEEE 802.15.4e Time-Slotted Channel Hopping (TSCH) deliver high reliability to fulfill the requirements in IIoT by following strict schedules computed in a Scheduling Function (SF) to avoid collisions and to provide determinism. The standard does not define how such schedules are built. The SF plays an essential role in 6TiSCH networks since it dictates when and where the nodes are communicating according to the application requirements, thus directly influencing the reliability of the network. Moreover, typical industrial environments consist of heavy machinery and complementary wireless communication systems that can create interference. Hence, we propose a distributed SF, namely the Channel Ranking Scheduling Function (CRSF), for IIoT networks supporting IPv6 over the IEEE 802.15.4e TSCH mode. CRSF computes the number of cells required for each node using a buffer-based bandwidth allocation mechanism with a Kalman filtering technique to avoid sudden allocation/deallocation of cells. CRSF also ranks channel quality using Exponential Weighted Moving Averages (EWMAs) based on the Received Signal Strength Indicator (RSSI), Background Noise (BN) level measurements, and the Packet Delivery Rate (PDR) metrics to select the best available channel to communicate. We compare the performance of CRSF with Orchestra and the Minimal Scheduling Function (MSF), in scenarios resembling industrial environmental characteristics. Performance is evaluated in terms of PDR, end-to-end latency, Radio Duty Cycle (RDC), and the elapsed time of first packet arrival. Results show that CRSF achieves high PDR and low RDC across all scenarios with periodic and burst traffic patterns at the cost of increased end-to-end latency. Moreover, CRSF delivers the first packet earlier than Orchestra and MSF in all scenarios. We conclude that CRSF is a viable option for IIoT networks with a large number of nodes and interference. The main contributions of our paper are threefold: (i) a bandwidth allocation mechanism that uses Kalman filtering techniques to effectively calculate the number of cells required for a given time, (ii) a channel ranking mechanism that combines metrics such as the PDR, RSSI, and BN to select channels with the best performance, and (iii) a new Key Performance Indicator (KPI) that measures the elapsed time from network formation until the first packet reception at the root.
ABSTRACT
Alternative wireless data communication systems are a necessity in industries that operate in harsh environments such as the oil and gas industry. Ultrasonic guided wave propagation through solid metallic structures, such as metal barriers, rods, and multiwire cables, have been proposed for data transmission purposes. In this context, multiwire cables have been explored as a communication media for the transmission of encoded ultrasonic guided waves. This work presents the proprietary hardware design and implementation of an automatic data transmission system based on the propagation of ultrasonic guided waves using as communication channels a high-temperature and corrosion-resistant oil industry multiwire cable. A dedicated communication protocol has been implemented at physical and data link layers, which involved pulse position modulation (PPM), digital signal processing (DSP), and an integrity validation byte. The data transmission system was composed of an ultrasonic guided waves PPM encoded data transmitter, a 1K22 MP-35N multiwire cable, a hardware preamplifier, a data acquisition module, a real-time (RT) DSP LabVIEW (National Instruments, Austin, TX) based demodulator, and a human-machine interface (HMI) running on a personal computer. To evaluate the communication system, the transmitter generated 60 kHz PPM energy packets containing three different bytes and their corresponding integrity validation bytes. Experimental tests were conducted in the laboratory using 1 and 10 m length cables. Although a dispersive solid elastic media was used as a communication channel, results showed that digital data transmission rates, up to 470 bps, were effectively validated.
