Electrical and Electronic Engineering - Conference Items

Permanent URI for this collection


Recent Submissions

Now showing 1 - 5 of 153
  • Item
    Neonatal hypoxic-ischemic encephalopathy grading from multi-channel EEG time-series data using a fully convolutional neural network
    (Institute of Electrical and Electronics Engineers (IEEE), 2023-10-18) Yu, Shuwen; Marnane, William P.; Boylan, Geraldine B.; Lightbody, Gordon; Science Foundation Ireland; Wellcome Trust
    A deep learning classifier is proposed for hypoxic-ischemic encephalopathy (HIE) grading in neonates. Rather than using any features, this architecture can be fed with raw EEG. Fully convolutional layers were adopted both in the feature extraction and classification blocks, which makes this architecture simpler, and deeper, but with fewer parameters. Here two large (335h and 338h respectively) multi-center neonatal continuous EEG datasets were used for training and test. The model was trained based on weak labels and channel independence. A majority vote method was used for the post-processing of the classifier results (across time and channels) to increase the robustness of the prediction. The proposed system achieved an accuracy of 86.09% (95% confidence interval: 82.41% ∼89.78%), an MCC of 0.7691, and an AUC of 86.23% on the large unseen test set. Two convolutional neural network architectures which utilized time-frequency distribution features were selected as the baseline as they had been developed or tested on the same datasets. A relative improvement of 23.65% in test accuracy was obtained as compared with the best baseline.
  • Item
    American Multinomial Option Pricing on FPGA using OneAPI
    (IEEE, 2023) O'Mahony, Aidan T.; Zeidan, Gil; Hanzon, Bernard; Popovici, Emanuel M.; Dell Technologies; Intel Corporation; Science Foundation Ireland
    This paper describes a new method for pricing American options that utilizes the OneAPI framework and a Field-Programmable Gate Array (FPGA) device. By using a multinomial model based on Pascals Simplex, the method is able to price American options with improved performance over traditional methods. The use of the OneAPI framework allows for efficient parallelization of the computations on the FPGA, resulting in improved performance and faster pricing of the options. Our analysis indicates that for 7 assets, the method can process approximately 53.74 options per second at the maximum tree depth of 15, showcasing its potential for real-world applications. The results of the method are validated against existing pricing models, demonstrating its accuracy and viability for use in real-world option pricing applications.
  • Item
    Behavioral modeling of low-frequency noise in switched-capacitor circuits using Python
    (Institute of Electrical and Electronics Engineers (IEEE), 2022-08-05) Kalogiros, Spyridon; Salgado, Gerardo; McCarthy, Kevin; O'Connell, Ivan; Science Foundation Ireland; European Regional Development Fund
    In precision circuits validating the performance in the presence of low-frequency noise is particularly challenging especially at transistor level, as long simulations are required to observe the low frequency performance. However, running such system-level simulations is rarely practical at transistor level as these simulations can take days to weeks to complete. This work presents a high-level model in Python for generating low-frequency noise which can be used for validating the low-frequency performance of a design in a timely manner. Simulation times can be reduced from days to minutes, enabling designers to achieve a high-level simulation coverage. With Python and NumPy this can be achieved using open-source software tools at no cost.
  • Item
    Lightweight anomaly detection framework for IoT
    (Institute of Electrical and Electronics Engineers (IEEE), 2020-08-31) Tagliaro Beasley, Bianca; O'Mahony, George D.; Gómez Quintana, Sergi; Temko, Andriy; Popovici, Emanuel; Science Foundation Ireland
    Internet of Things (IoT) security is growing in importance in many applications ranging from biomedical to environmental to industrial applications. Access to data is the primary target for many of these applications. Often IoT devices are an essential part of critical control systems that could affect well-being, safety, or inflict severe financial damage. No current solution addresses all security aspects. This is mainly due to the resource-constrained nature of IoT, cost, and power consumption. In this paper, we propose and analyse a framework for detecting anomalies on a low power IoT platform. By monitoring power consumption and by using machine learning techniques, we show that we can detect a large number and types of anomalies during the execution phase of an application running on the IoT. The proposed methodology is generic in nature, hence allowing for deployment in a myriad of scenarios.
  • Item
    FPGA hardware acceleration framework for anomaly-based intrusion detection system in IoT
    (Institute of Electrical and Electronics Engineers (IEEE), 2021-10-12) Ngo, Duc-Minh; Temko, Andriy; Murphy, Colin C.; Popovici, Emanuel; Science Foundation Ireland; European Regional Development Fund
    This study proposes a versatile framework for realtime Internet of Things (IoT) network intrusion detection using Artificial Neural Network (ANN) on heterogeneous hardware. With the increase in the volume of exchanged data, IoT networks' security has become a crucial issue. Anomaly-based intrusion detection systems (IDS) using machine learning have recently gained increased popularity due to their generation ability to detect new attacks. However, the deployment of anomaly-based AI-assisted IDS for IoT devices is computationally expensive. In this paper, a hierarchical decision-making approach for IDS is proposed and evaluated on the new IoT-23 dataset, with improved accuracy over the software-based methods. The inference engine is implemented on the Xilinx FPGA System on a Chip (SoC) hardware platform for high performance, high accuracy attack detection (more than 99.43%). For the resulting implemented design, the processing time of the ANN model on FPGA with an xc7z020clg400 device is 6.6 times and 40.5 times faster than GPU Quadro M2000 and CPU E5-2640 2.60GHz, respectively.