Search

In the modern world, connectivity and shared intelligence enable independent constituent systems (CS) to form systems of systems (SoS) capable of performing sophisticated missions. However, the sheer scale of an SoS can make it challenging to manage all components comprehensively, hiding potential security and safety concerns. These factors underscore the need for advancing conceptual models that permit a better understanding of the SoS intricacies. This paper presents a conceptual model for an integrated safety-security ontology for SoS, called SSO-SoS. Such a model is based on international standards, existing literature, and relevant conceptual models, where we pay special attention to safety, security, and mitigation for SoS. We also illustrate the SSO-SoS with a case study from the construction sector. Our conceptual model provides a hierarchical organization that permits stakeholders to navigate through different layers of information, enhancing their ability to identify, address, and understand the required SoS knowledge.

Assemblies of machinery commonly require control systems whose functionality is based on application software. In Europe, such software requires high safety integrity levels in accordance with the Machinery Directive (MD). However, identifying the essential regulatory requirements for the safety approval is not an easy task. To facilitate this job, this paper presents a process for Safety Argumentation for Machinery Assembly Control Software (SAMACS). We are inspired by patterns provided in the Goal Structuring Notation (GSN) and the use of contracts in safety argumentation. SAMACS contribution is aligning those methods with the MD by adopting EN ISO 13849. In particular, we define safety goals based on expected software contribution to control system safety and the standard guidance. Software safety goals are detailed into software safety requirements and expressed further as contracts, which shall be verified with prescribed techniques. We apply SAMACS to a case study from a European mining company and discuss the findings. This work aims at helping practitioners compose the safety case argumentation necessary to support machinery integration approval in Europe.

CAMP proposes a hierarchical cache subsystem for multi-core mixed criticality processors, focusing on ensuring worst-case execution time (WCET) predictability in automotive applications. It incorporates criticality-aware locked L1 and L2 caches, reconfigurable at mode change intervals, along with criticality-aware last level cache partitioning. Evaluation using CACOSIM, Moola Multicore simulator, and CACTI simulation tools confirms the suitability of CAMP for keeping high-criticality jobs within timing budgets. A practical case study involving an automotive wake-up controller using the sniper v7.2 architecture simulator further validates its usability in real-world mixed criticality applications. CAMP presents a promising cache architecture for optimized multi-core mixed criticality systems.

The past decade has seen tremendous interest in mixed criticality systems research due to its exponential growth with inherent challenges of effective resource utilisation and isolation. The pervasiveness of these systems along with their certification needs, prompt for suitable task models to perform the required analysis. Extensive usage scenarios and strict certification requirements have spawned a broad spectrum of research and evolved into several task models. In this work, a thematic survey of task models for both uni-core and multi-core mixed criticality systems is carried out. The work categorises task models based on attributes such as resources, quality of service, operating system overheads, energy, fault tolerance and parallel processing. After synthesising the state-of-the-art, the work summarises task models by providing a visual aid and a ready reckoner with traceability to mixed criticality challenges. This work serves as a quintessential reference manual for researchers and academicians in the mixed criticality domain.

Hazard analysis for a single system focuses on identifying and evaluating potential hazards associated with the individual system, its components, and their interactions. There are well-established hazard analysis techniques that are widely used to identify hazards for single systems. However, unlike single systems, hazard analysis in a System of Systems (SoS) must focus on analyzing the potential hazards (including emergent ones) that can arise from the interactions between multiple individual systems. This type of analysis considers the complex interactions between systems and the interdependence between their components and the environment in which they operate. Therefore, it is necessary to understand the application scenarios of SoS and to employ a systematic approach to identify all potential hazards. This paper applies the Composite Hazard Analysis Technique (CompHAT) to an industrial case study from a mining and equipment domain. The results show that the CompHAT is useful in identifying the interaction faults and their propagation routes between components of a constituent system and between constituent systems in an SoS. We also report that, due to the tool support, CompHAT can be beneficial for safety engineers to trace the faults in the network of an SoS in a more efficient and effective manner.

Modern manufacturing systems collect a huge amount of data which gives an opportunity to apply various Machine Learning (ML) techniques. The focus of this paper is on the detection of anomalous behavior in industrial manufacturing systems by considering the temporal nature of the manufacturing process. Long Short-Term Memory (LSTM) networks are applied on a publicly available dataset called Modular Ice-cream factory Dataset on Anomalies in Sensors (MIDAS), which is created using a simulation of a modular manufacturing system for ice cream production. Two different problems are addressed: anomaly detection and anomaly classification. LSTM performance is analysed in terms of accuracy, execution time, and memory consumption and compared with non-time-series ML algorithms including Logistic Regression, Decision Tree, Random Forest, and Multi-Layer Perceptron. The experiments demonstrate the importance of considering the temporal nature of the manufacturing process in detecting anomalous behavior and the superiority in accuracy of LSTM over non-time-series ML algorithms. Additionally, runtime adaptation of the predictions produced by LSTM is proposed to enhance its applicability in a real system.

Cyber-physical systems that combine sensing, computing, control and networking with physical items and infrastructure, such as automotive, avionics and robotics, are rapidly becoming mixed criticality systems (MCS). The increasing expectations for computing ability and predictable temporal behaviour of these systems necessitate substantial enhancements in their memory subsystem architecture. The use of locked caches to have predictable execution time is one such optimization. There is no comprehensive method in order to manage coherency in locked caches in any of the current cache coherence protocols like MOESI. CLAMP—A criticality aware coherency protocol for locked multi-level caches in multi-core processors is an updated variant of MOESI and as an extension of MOESIL, to improve the data consistency of locked caches. The work CLAMP proposes an improvised locked cache coherence protocol for multiple levels of cache in multi-core MCS, whereas MOESIL is restricted to two-level cache architecture. Experiments using real-time benchmark programs on CACOSIM reveal an average cache miss rate reduction of 18% for high-criticality jobs.

Tight end-to-end worst-case delay bounds for periodic traffic streams are essential for time sensitive networks. In this paper, we provide an algorithm to compute a tight (and accurate) end-to-end worst-case bound by considering distinct topological patterns and the manner in which streams enter and leave switches. This refined analysis uses non-preemptive, strict-priority arbitration mechanism commonly deployed in Ethernet switches. Compared to the state-of-the-art that considers all higher and equal priority interference as contributing to the worst-case bound, we present an analytical approach for computing a tighter worst-case delay bound and prove through discrete event simulations that only a certain number of equal-priority interference streams can actually affect the worst-case case. Our results enable efficient resource allocation and have implications for online re-configuration mechanisms for time-sensitive factory communication systems.

Tiny machine learning (TinyML) represents an emerging research direction that aims to realize machine learning on Internet of Things (IoT) devices. The current TinyML research seems to focus on supporting the deployment of deep learning models on microprocessors, while the models themselves are trained on high performance computers or clouds. However, in the resource/time constrained IoT contexts, it is more desirable to perform data analytics and learning tasks directly on edge devices for crucial benefits such as increased energy efficiency, reduced latency as well as lower communication cost.To address the above challenge, this paper proposes a tiny federated learning algorithm for enabling learning of Bayesian classifiers based on distributed tiny data storage, referred to as TFL-BC. In TFL-BC, Bayesian learning is executed in parallel across multiple edge devices using local (tiny) training data and subsequently the learning results from local devices are aggregated via a central node to obtain the final classification model. The results of experiments conducted on a set of benchmark datasets demonstrate that our algorithm can produce final aggregated models that outperform single tiny Bayesian classifiers and that the result of tiny federated learning (of Bayesian classifier) is independent of the number of data partitions used for generating the distributed local training data.

There is a growing body of knowledge on network intrusion detection, and several open data sets with network traffic and cyber-security threats have been released in the past decades. However, many data sets have aged, were not collected in a contemporary industrial communication system, or do not easily support research focusing on distributed anomaly detection. This paper presents the Westermo network traffic data set, 1.8 million network packets recorded in over 90 minutes in a network built up of twelve hardware devices. In addition to the raw data in PCAP format, the data set also contains pre-processed data in the form of network flows in CSV files. This data set can support the research community for topics such as intrusion detection, anomaly detection, misconfiguration detection, distributed or federated artificial intelligence, and attack classification. In particular, we aim to use the data set to continue work on resource-constrained distributed artificial intelligence in edge devices. The data set contains six types of events: harmless SSH, bad SSH, misconfigured IP address, duplicated IP address, port scan, and man in the middle attack.

There is a growing number of network attacks and the data on the network is more exposed than ever with the increased activity on the Internet. Applying Machine Learning (ML) techniques for cyber-security is a popular and effective approach to address this problem. However, the data which is used by ML algorithms have to be protected. In this paper, we present a framework that combines autoencoder, multi-objective optimization algorithms, and different ML algorithms to encode the network data, reduce its size, and detect and classify the network attacks. The novel element used in this framework, with respect to earlier research, is the application of multi-objective optimization algorithms, such as Multi-Objective Differential Evolution or Non-dominated Sorting Genetic Algorithm-II, to handle the different objectives in the fitness function of the autoencoder (autoencoder decoding error and accuracy of ML algorithm). We evaluated six different ML algorithms for attack detection and classification on network dataset UNSWNB15. The performance of the proposed framework is compared with single-objective Differential Evolution. The results showed that Multi-Objective Differential Evolution outperforms the counterparts for attack detection, while all the evaluated algorithms showed similar performance for attack classification.

Multi-core architectures have grown to be a popular choice for deploying Mixed Criticality Systems (MCS). The focus of research in MCS has been to provide timing assurances for jobs with different criticality levels. Due to their significant processing demands and energy-aware/constrained nature, energy conservation in these systems is becoming mandatory. This article presents, mcDVFS, an energy management technique based on Dynamic-Voltage-and-Frequency-Scaling for multi-core MCS. mcDVFS achieves significant energy reduction while maintaining timing guarantees. It also prioritizes quality of service whenever feasible. Extensive experimental simulations show energy savings of ≈ 52% and 34% when compared to EDF-VD and EDF-VD with QoS.