arXiv subjects

Robert Atkinson

Publications and source records attributed to Robert Atkinson.

At least 19 recordsLinked to original sources

arXiv2026

Few-Shot Network Intrusion Detection Using Online Triplet Mining

Network intrusion detection systems play a vital role in protecting networks by detecting malicious network traffic which can then be investigated by a cybersecurity operations centre. State-of-the-art approaches utilise supervised machine learning methods to train a classification model to recognise known cyberattacks; however, these models require a large labelled dataset to train and show poor performance when trained on smaller datasets. In an attempt to address this shortcoming, anomaly detection models learn the distribution of benign traffic and flag non-conforming traffic as malicious. While these methods do not require malicious examples to train, they suffer from high false-positive rates rendering them impractical. As a result, networks may be particularly vulnerable when there are insufficient labelled instances of a specific attack class to train an effective classifier. This often occurs in newly established networks or when previously unseen types of attacks emerge. To address this challenge, this work proposes the use of a triplet network, utilising online triplet mining and a KNN classifier, which is able to perform few-shot classification, enabling effective intrusion detection after being trained on a limited number of malicious examples. Various online triplet mining algorithms were explored and model design choices, such as the inference algorithm and optimised distance metrics, were compared and evaluated through a series of ablation studies. The final model was compared against other state-of-the-art approaches in few-shot binary and multiclass classification, where the proposed approach was found to be competitive with existing methods when trained on as little as 10 malicious samples of each class.

cs.CR↗

arXiv2026

A Novel Contrastive Loss for Zero-Day Network Intrusion Detection

Machine learning has achieved state-of-the-art results in network intrusion detection; however, its performance significantly degrades when confronted by a new attack class -- a zero-day attack. In simple terms, classical machine learning-based approaches are adept at identifying attack classes on which they have been previously trained, but struggle with those not included in their training data. One approach to addressing this shortcoming is to utilise anomaly detectors which train exclusively on benign data with the goal of generalising to all attack classes -- both known and zero-day. However, this comes at the expense of a prohibitively high false positive rate. This work proposes a novel contrastive loss function which is able to maintain the advantages of other contrastive learning-based approaches (robustness to imbalanced data) but can also generalise to zero-day attacks. Unlike anomaly detectors, this model learns the distributions of benign traffic using both benign and known malign samples, i.e. other well-known attack classes (not including the zero-day class), and consequently, achieves significant performance improvements. The proposed approach is experimentally verified on the Lycos2017 dataset where it achieves an AUROC improvement of .000065 and .060883 over previous models in known and zero-day attack detection, respectively. Finally, the proposed method is extended to open-set recognition achieving OpenAUC improvements of .170883 over existing approaches.

cs.CR↗

arXiv2025

Coherent Load Profile Synthesis with Conditional Diffusion for LV Distribution Network Scenario Generation

Limited visibility of distribution network power flows at the low voltage level presents challenges to both distribution network operators from a planning perspective and distribution system operators from a congestion management perspective. More representative loads are required to support meaningful analysis of LV substations; otherwise, such analysis risks misinforming future decisions. Traditional load profiling relies on typical profiles, oversimplifying substation-level complexity. Generative models have attempted to address this through synthesising representative loads from historical exemplars; however, while these approaches can approximate load shapes to a convincing degree of fidelity, analysis of the co-behaviour between substations is limited, which ultimately impacts higher voltage level network operation. This limitation will become even more pronounced with the increasing integration of low-carbon technologies, as estimates of base loads fail to capture load diversity. To address this gap, Conditional Diffusion models for synthesising daily active and reactive power profiles at the low voltage distribution substation level are proposed. The evaluation of fidelity is demonstrated through conventional metrics capturing temporal and statistical realism, as well as power flow modelling. Multiple models are proposed to handle varying levels of data availability, ranging from unconditional synthesis to an informed generation driven by metadata and daily statistics. The results show synthesised load profiles are plausible both independently and as a cohort in a wider power systems context. The Conditional Diffusion model is benchmarked against both naive and state-of-the-art models to demonstrate its effectiveness in producing realistic scenarios on which to base sub-regional power distribution network planning and operations.

eess.SY↗

arXiv2025

Signal-Based Malware Classification Using 1D CNNs

Malware classification is a contemporary and ongoing challenge in cyber-security: modern obfuscation techniques are able to evade traditional static analysis, while dynamic analysis is too resource intensive to be deployed at a large scale. One prominent line of research addresses these limitations by converting malware binaries into 2D images by heuristically reshaping them into a 2D grid before resizing using Lanczos resampling. These images can then be classified based on their textural information using computer vision approaches. While this approach can detect obfuscated malware more effectively than static analysis, the process of converting files into 2D images results in significant information loss due to both quantisation noise, caused by rounding to integer pixel values, and the introduction of 2D dependencies which do not exist in the original data. This loss of signal limits the classification performance of the downstream model. This work addresses these weaknesses by instead resizing the files into 1D signals which avoids the need for heuristic reshaping, and additionally these signals do not suffer from quantisation noise due to being stored in a floating-point format. It is shown that existing 2D CNN architectures can be readily adapted to classify these 1D signals for improved performance. Furthermore, a bespoke 1D convolutional neural network, based on the ResNet architecture and squeeze-and-excitation layers, was developed to classify these signals and evaluated on the MalNet dataset. It was found to achieve state-of-the-art performance on binary, type, and family level classification with F1 scores of 0.874, 0.503, and 0.507, respectively, paving the way for future models to operate on the proposed signal modality.

cs.CR↗

arXiv2025

Contrastive Self-Supervised Network Intrusion Detection using Augmented Negative Pairs

Network intrusion detection remains a critical challenge in cybersecurity. While supervised machine learning models achieve state-of-the-art performance, their reliance on large labelled datasets makes them impractical for many real-world applications. Anomaly detection methods, which train exclusively on benign traffic to identify malicious activity, suffer from high false positive rates, limiting their usability. Recently, self-supervised learning techniques have demonstrated improved performance with lower false positive rates by learning discriminative latent representations of benign traffic. In particular, contrastive self-supervised models achieve this by minimizing the distance between similar (positive) views of benign traffic while maximizing it between dissimilar (negative) views. Existing approaches generate positive views through data augmentation and treat other samples as negative. In contrast, this work introduces Contrastive Learning using Augmented Negative pairs (CLAN), a novel paradigm for network intrusion detection where augmented samples are treated as negative views - representing potentially malicious distributions - while other benign samples serve as positive views. This approach enhances both classification accuracy and inference efficiency after pretraining on benign traffic. Experimental evaluation on the Lycos2017 dataset demonstrates that the proposed method surpasses existing self-supervised and anomaly detection techniques in a binary classification task. Furthermore, when fine-tuned on a limited labelled dataset, the proposed approach achieves superior multi-class classification performance compared to existing self-supervised models.

cs.LG↗

arXiv2024

Non-invasive Diver Respiration Rate Monitoring in Hyperbaric Lifeboat Environments using Short-Range Radar

The monitoring of diver health during emergency events is crucial to ensuring the safety of personnel. A non-invasive system continuously providing a measure of the respiration rate of individual divers is exceedingly beneficial in this context. The paper reports on the application of short-range radar to record the respiration rate of divers within hyperbaric lifeboat environments. Results demonstrate that the respiratory motion can be extracted from the radar return signal applying routine signal processing. Further, evidence is provided that the radar-based approach yields a more accurate measure of respiration rate than an audio signal from a headset microphone. The system promotes an improvement in evacuation protocols under critical operational scenarios.

eess.SP↗

arXiv2023

Detecting Cloud Presence in Satellite Images Using the RGB-based CLIP Vision-Language Model

This work explores capabilities of the pre-trained CLIP vision-language model to identify satellite images affected by clouds. Several approaches to using the model to perform cloud presence detection are proposed and evaluated, including a purely zero-shot operation with text prompts and several fine-tuning approaches. Furthermore, the transferability of the methods across different datasets and sensor types (Sentinel-2 and Landsat-8) is tested. The results that CLIP can achieve non-trivial performance on the cloud presence detection task with apparent capability to generalise across sensing modalities and sensing bands. It is also found that a low-cost fine-tuning stage leads to a strong increase in true negative rate. The results demonstrate that the representations learned by the CLIP model can be useful for satellite image processing tasks involving clouds.

cs.CV↗

arXiv2021

Neural Weight Step Video Compression

A variety of compression methods based on encoding images as weights of a neural network have been recently proposed. Yet, the potential of similar approaches for video compression remains unexplored. In this work, we suggest a set of experiments for testing the feasibility of compressing video using two architectural paradigms, coordinate-based MLP (CbMLP) and convolutional network. Furthermore, we propose a novel technique of neural weight stepping, where subsequent frames of a video are encoded as low-entropy parameter updates. To assess the feasibility of the considered approaches, we will test the video compression performance on several high-resolution video datasets and compare against existing conventional and neural compression techniques.

cs.CV↗

arXiv2021

Neural Knitworks: Patched Neural Implicit Representation Networks

Coordinate-based Multilayer Perceptron (MLP) networks, despite being capable of learning neural implicit representations, are not performant for internal image synthesis applications. Convolutional Neural Networks (CNNs) are typically used instead for a variety of internal generative tasks, at the cost of a larger model. We propose Neural Knitwork, an architecture for neural implicit representation learning of natural images that achieves image synthesis by optimizing the distribution of image patches in an adversarial manner and by enforcing consistency between the patch predictions. To the best of our knowledge, this is the first implementation of a coordinate-based MLP tailored for synthesis tasks such as image inpainting, super-resolution, and denoising. We demonstrate the utility of the proposed technique by training on these three tasks. The results show that modeling natural images using patches, rather than pixels, produces results of higher fidelity. The resulting model requires 80% fewer parameters than alternative CNN-based solutions while achieving comparable performance and training time.

cs.CV↗

arXiv2021

Cyber-Security Challenges in Aviation Industry: A Review of Current and Future Trends

The integration of Information and Communication Technology (ICT) tools into mechanical devices found in aviation industry has raised security concerns. The more integrated the system, the more vulnerable due to the inherent vulnerabilities found in ICT tools and software that drives the system. The security concerns have become more heightened as the concept of electronic-enabled aircraft and smart airports get refined and implemented underway. In line with the above, this paper undertakes a review of cyber-security incidence in the aviation sector over the last 20 years. The essence is to understand the common threat actors, their motivations, the type of attacks, aviation infrastructure that is commonly attacked and then match these so as to provide insight on the current state of the cyber-security in the aviation sector. The review showed that the industry's threats come mainly from Advance Persistent Threat (APT) groups that work in collaboration with some state actors to steal intellectual property and intelligence, in order to advance their domestic aerospace capabilities as well as possibly monitor, infiltrate and subvert other nations' capabilities. The segment of the aviation industry commonly attacked is the Information Technology infrastructure, and the prominent type of attacks is malicious hacking activities that aim at gaining unauthorised access using known malicious password cracking techniques such as Brute force attacks, Dictionary attacks and so on. The review further analysed the different attack surfaces that exist in aviation industry, threat dynamics, and use these dynamics to predict future trends of cyberattacks in the industry. The aim is to provide information for the cybersecurity professionals and aviation stakeholders for proactive actions in protecting these critical infrastructures against cyberincidence for an optimal customer service oriented industry.

cs.CR↗

arXiv2021

Utilising Flow Aggregation to Classify Benign Imitating Attacks

Cyber-attacks continue to grow, both in terms of volume and sophistication. This is aided by an increase in available computational power, expanding attack surfaces, and advancements in the human understanding of how to make attacks undetectable. Unsurprisingly, machine learning is utilised to defend against these attacks. In many applications, the choice of features is more important than the choice of model. A range of studies have, with varying degrees of success, attempted to discriminate between benign traffic and well-known cyber-attacks. The features used in these studies are broadly similar and have demonstrated their effectiveness in situations where cyber-attacks do not imitate benign behaviour. To overcome this barrier, in this manuscript, we introduce new features based on a higher level of abstraction of network traffic. Specifically, we perform flow aggregation by grouping flows with similarities. This additional level of feature abstraction benefits from cumulative information, thus qualifying the models to classify cyber-attacks that mimic benign traffic. The performance of the new features is evaluated using the benchmark CICIDS2017 dataset, and the results demonstrate their validity and effectiveness. This novel proposal will improve the detection accuracy of cyber-attacks and also build towards a new direction of feature extraction for complex ones.

cs.CR↗

arXiv2020

A Review of Cyber-Ranges and Test-Beds: Current and Future Trends

Cyber situational awareness has been proven to be of value in forming a comprehensive understanding of threats and vulnerabilities within organisations, as the degree of exposure is governed by the prevailing levels of cyber-hygiene and established processes. A more accurate assessment of the security provision informs on the most vulnerable environments that necessitate more diligent management. The rapid proliferation in the automation of cyber-attacks is reducing the gap between information and operational technologies and the need to review the current levels of robustness against new sophisticated cyber-attacks, trends, technologies and mitigation countermeasures has become pressing. A deeper characterisation is also the basis with which to predict future vulnerabilities in turn guiding the most appropriate deployment technologies. Thus, refreshing established practices and the scope of the training to support the decision making of users and operators. The foundation of the training provision is the use of Cyber-Ranges (CRs) and Test-Beds (TBs), platforms/tools that help inculcate a deeper understanding of the evolution of an attack and the methodology to deploy the most impactful countermeasures to arrest breaches. In this paper, an evaluation of documented CR and TB platforms is evaluated. CRs and TBs are segmented by type, technology, threat scenarios, applications and the scope of attainable training. To enrich the analysis of documented CR and TB research and cap the study, a taxonomy is developed to provide a broader comprehension of the future of CRs and TBs. The taxonomy elaborates on the CRs/TBs different dimensions, as well as, highlighting a diminishing differentiation between application areas.

cs.CR↗

arXiv2020

Interoperability and Integration Testing Methods for IoT Systems: a Systematic Mapping Study

The recent active development of Internet of Things (IoT) solutions in various domains has led to an increased demand for security, safety, and reliability of these systems. Security and data privacy are currently the most frequently discussed topics; however, other reliability aspects also need to be focused on to maintain the smooth and safe operation of IoT systems. Until now, there has been no systematic mapping study dedicated to the topic of interoperability and integration testing of IoT systems specifically; therefore, we present such an overview in this study. We analyze 803 papers from four major primary databases and perform detailed assessment and quality check to find 115 relevant papers. In addition, recently published testing techniques and approaches are analyzed and classified; the challenges and limitations in the field is also identified and discussed. Research trends related to publication time, active researchers, and publication media are presented in this study. The results suggest that studies mainly focus only on general testing methods, which can be applied to integration and interoperability testing of IoT systems; thus, there are research opportunities to develop additional testing methods focused specifically on IoT systems, so that they are more effective in the IoT context.

cs.SE↗

arXiv2020

Machine Learning Based IoT Intrusion Detection System: An MQTT Case Study (MQTT-IoT-IDS2020 Dataset)

The Internet of Things (IoT) is one of the main research fields in the Cybersecurity domain. This is due to (a) the increased dependency on automated device, and (b) the inadequacy of general purpose Intrusion Detection Systems (IDS) to be deployed for special purpose networks usage. Numerous lightweight protocols are being proposed for IoT devices communication usage. One of the distinguishable IoT machine-to-machine communication protocols is Message Queuing Telemetry Transport (MQTT) protocol. However, as per the authors best knowledge, there are no available IDS datasets that include MQTT benign or attack instances and thus, no IDS experimental results available. In this paper, the effectiveness of six Machine Learning (ML) techniques to detect MQTT-based attacks is evaluated. Three abstraction levels of features are assessed, namely, packet-based, unidirectional flow, and bidirectional flow features. An MQTT simulated dataset is generated and used for the training and evaluation processes. The dataset is released with an open access licence to help the research community further analyse the accompanied challenges. The experimental results demonstrated the adequacy of the proposed ML models to suit MQTT-based networks IDS requirements. Moreover, the results emphasise on the importance of using flow-based features to discriminate MQTT-based attacks from benign traffic, while packet-based features are sufficient for traditional networking attacks.

cs.CR↗

arXiv2020

Leveraging Siamese Networks for One-Shot Intrusion Detection Model

The use of supervised Machine Learning (ML) to enhance Intrusion Detection Systems has been the subject of significant research. Supervised ML is based upon learning by example, demanding significant volumes of representative instances for effective training and the need to re-train the model for every unseen cyber-attack class. However, retraining the models in-situ renders the network susceptible to attacks owing to the time-window required to acquire a sufficient volume of data. Although anomaly detection systems provide a coarse-grained defence against unseen attacks, these approaches are significantly less accurate and suffer from high false-positive rates. Here, a complementary approach referred to as 'One-Shot Learning', whereby a limited number of examples of a new attack-class is used to identify a new attack-class (out of many) is detailed. The model grants a new cyber-attack classification without retraining. A Siamese Network is trained to differentiate between classes based on pairs similarities, rather than features, allowing to identify new and previously unseen attacks. The performance of a pre-trained model to classify attack-classes based only on one example is evaluated using three datasets. Results confirm the adaptability of the model in classifying unseen attacks and the trade-off between performance and the need for distinctive class representation.

cs.CR↗

arXiv2020

Utilising Deep Learning Techniques for Effective Zero-Day Attack Detection

Machine Learning (ML) and Deep Learning (DL) have been used for building Intrusion Detection Systems (IDS). The increase in both the number and sheer variety of new cyber-attacks poses a tremendous challenge for IDS solutions that rely on a database of historical attack signatures. Therefore, the industrial pull for robust IDSs that are capable of flagging zero-day attacks is growing. Current outlier-based zero-day detection research suffers from high false-negative rates, thus limiting their practical use and performance. This paper proposes an autoencoder implementation for detecting zero-day attacks. The aim is to build an IDS model with high recall while keeping the miss rate (false-negatives) to an acceptable minimum. Two well-known IDS datasets are used for evaluation-CICIDS2017 and NSL-KDD. In order to demonstrate the efficacy of our model, we compare its results against a One-Class Support Vector Machine (SVM). The manuscript highlights the performance of a One-Class SVM when zero-day attacks are distinctive from normal behaviour. The proposed model benefits greatly from autoencoders encoding-decoding capabilities. The results show that autoencoders are well-suited at detecting complex zero-day attacks. The results demonstrate a zero-day detection accuracy of [89-99%] for the NSL-KDD dataset and [75-98%] for the CICIDS2017 dataset. Finally, the paper outlines the observed trade-off between recall and fallout.

cs.CR↗

arXiv2019

A Security Perspective on Unikernels

Cloud-based infrastructures have grown in popularity over the last decade leveraging virtualisation, server, storage, compute power and network components to develop flexible applications. The requirements for instantaneous deployment and reduced costs have led the shift from virtual machine deployment to containerisation, increasing the overall flexibility of applications and increasing performances. However, containers require a fully fleshed operating system to execute, increasing the attack surface of an application. Unikernels, on the other hand, provide a lightweight memory footprint, ease of application packaging and reduced start-up times. Moreover, Unikernels reduce the attack surface due to the self-contained environment only enabling low-level features. In this work, we provide an exhaustive description of the unikernel ecosystem; we demonstrate unikernel vulnerabilities and further discuss the security implications of Unikernel-enabled environments through different use-cases.

cs.CR↗

arXiv2019

Cyber-Security Internals of a Skoda Octavia vRS: A Hands on Approach

The convergence of information technology and vehicular technologies are a growing paradigm, allowing information to be sent by and to vehicles. This information can further be processed by the Electronic Control Unit (ECU) and the Controller Area Network (CAN) for in-vehicle communications or through a mobile phone or server for out-vehicle communication. Information sent by or to the vehicle can be life-critical (e.g. breaking, acceleration, cruise control, emergency communication, etc. . . ). As vehicular technology advances, in-vehicle networks are connected to external networks through 3 and 4G mobile networks, enabling manufacturer and customer monitoring of different aspects of the car. While these services provide valuable information, they also increase the attack surface of the vehicle, and can enable long and short range attacks. In this manuscript, we evaluate the security of the 2017 Skoda Octavia vRS 4x4. Both physical and remote attacks are considered, the key fob rolling code is successfully compromised, privacy attacks are demonstrated through the infotainment system, the Volkswagen Transport Protocol 2.0 is reverse engineered. Additionally, in-car attacks are highlighted and described, providing an overlook of potentially deadly threats by modifying ECU parameters and components enabling digital forensics investigation are identified.

cs.CR↗