Skip to main content

2025 | Buch

Internet of Things

7th IFIP WG 5.5 International Cross-Domain Conference, IFIPIoT 2024, Nice, France, November 6–8, 2024, Proceedings

insite
SUCHEN

Über dieses Buch

This book constitutes the refereed proceedings of the 7th IFIP WG 5.5 International Cross-Domain Conference on Internet of Things, IFIPIoT 2024, in Nice, France, in November 2024.737

The 13 full papers and 4 short papers presented were carefully reviewed and selected from a total of 28 submissions to the main conference. They were organized in topical sections as follows: Hardware/Software Solutions for IoT and CPS; Electronics and Signal Processing for IoT ; Networking and Communications Technology for IoT; Artificial Intelligence and Machine Learning Technologies for IoT; Cyber Security/Privacy/Trust for IoT and CPS and IoT or CPS Applications and Use cases.

Inhaltsverzeichnis

Frontmatter

Hardware/Software Solutions for IoT and CPS

Frontmatter
Relocation of Container-Based Services in a MEC-NFV Orchestrated Environment
Abstract
With the rapid growth of real-time next-generation mobile services, it has become necessary to work towards holistic orchestration of the benefits promised with edge computing based on bringing the computing infrastructure closer to the end user. While the concept of Multi-access Edge Computing (MEC) integrated with Network Function Virtualisation (NFV) is being standardised, there is still a lot of work to be done to orchestrate the relocation of edge applications integrated in 5G and beyond systems in a smooth and efficient manner. In this paper, we document the current status of the transparent relocation of edge services in an experimentally deployed MEC-NFV environment based on OSM. Working towards gathering monitoring training datasets necessary for the development of proactive MEC application orchestrators that will implement seamless follow-me behaviour for MEC services, we provide benchmark results for the service downtime of three potential MEC services hosted in lightweight containers. Our analysis of results shows that containers exhibit improved performance over that of virtual machines, but there are still some issues that require improvement in both the orchestration implementation as well at the relocation process for containers.
Cristina Bernad, Vojdan Kjorveziroski, Pedro Roig, Salvador Alcaraz, Katja Gilly, Sonja Filiposka
The Good, the Bad and the Ugly: Investigating the Effectiveness of Graph Deep Neural Networks for Anomaly Detection in Industrial Control Systems
Abstract
Industrial Control Systems (ICS) are paramount to the efficient operation of Critical National Infrastructure (CNI) ranging from electricity generation and distribution to manufacturing. However, the growing convergence of ICS with Information Technology (IT) systems renders CNI vulnerable to a range of cyber threats. Graph neural networks are being increasingly used for anomaly detection by adding granularity to the detection process. In this paper, we present a comparative study of graph-based deep learning models for ICS anomaly detection. Through the evaluation of four models using three multivariate industrial datasets, we aim to discern the effectiveness of prediction and reconstruction-based graph models in the ICS domain. We investigate data reduction techniques to minimise features needed to represent the window size and examine the representation of sliding window in terms of feature size for time-series analysis. Additionally, we assess the impact of the length of a context window on anomaly detection performance. Our results show that using feature reduction techniques on a longer context window produces better results while having the computational advantages of a shorter window size. Graph autoencoder is the most resilient to feature size reduction by maintaining similar F1 and AUC-PR score regardless of the number of features used to represent a context window. The results also provide insight to the suitability of graph-based models in this domain and offer recommendations for their optimal usage, paving the way for enhanced security and resilience in ICS.
Martin Nahalka, Marco M. Cook, Dimitrios Pezaros
Programmable and Scalable Bit-Sliced VLSI Architecture for Decision Tree-Based Machine Learning Edge Inference
Abstract
As the volume and diversity of Internet of Things (IoT) data continues to grow, traditional cloud-based processing methods face significant challenges, including latency, bandwidth constraints, and privacy concerns. Our research focuses on employing decision trees (DTs) [1] as an intelligent filtering mechanism on the edge. We propose a novel programmable and scalable custom ASIC architecture designed for decision tree based Machine Learning (ML) inference. Each bit-slice incorporates two 8-bit SISO input registers connected to an 8-bit comparator for data processing, the output of the comparator drives the select line of Mux, which selects the respective true and false paths. Each bit-slice can be programmed into either a leaf node or a regular node. A leaf node stores classification labels. A regular node compares a feature value with a weight value to decide between true and false paths. Given a DT model, the decision tree can be pre-programmed to store the model weights in respective tree nodes. In the inference phase, feature values are sequentially fed into the DT nodes. After the feature values are loaded the DT tree performs an inference with the classification value generated in the root node. We have implemented and validated the architecture at the layout level using Cadence Virtuoso in 0.5 \(\upmu \)m CMOS technology node. A 5-level DT occupies roughly 90 mm\(^2\) area with 22.58 mW of power consumption at a maximum clock speed of 12.8 MHz.
Raaga Sai Somesula, Sibi Rajagopal Sivakumar, Srinivas Katkoori

Electronics and Signal Processing for IoT

Frontmatter
Efficient Implementation of Authenticated Encryption on 16-bit MSP430 Microcontrollers
Abstract
Algorithms for Authenticated Encryption with Associated Data (AEAD) extend the normal functionality of authenticated encryption schemes by the ability to process data that is only authenticated but not encrypted. Such algorithms have attracted much interest in the past few years, especially the question of how they can be designed and implemented efficiently to perform well in resource-constrained devices like miniature sensor nodes or RFID tags. In this paper, we analyze the performance of the lightweight AEAD schemes Elephant v2, Grain-128AEADv2, ISAP v2.0, PHOTON-Beetle, and Romulus v1.3 on the MSP430 family of 16-bit ultra-low-power microcontrollers. All five have in common that they offer large security margins and made it into the last round of the Lightweight Cryptography (LWC) standardization project of the U.S. National Institute of Standards and Technology. We describe how these AEAD algorithms can be implemented efficiently in software and introduce Assembly-level optimization techniques for the underlying primitives, which include three permutations, one tweakable block cipher, and one stream cipher. Furthermore, we present numerous detailed benchmarking results (i.e., execution time and code size) for the primitives as well as for the full AEAD algorithms for different lengths of plaintext and associated data. Our benchmarks clearly show that all five AEAD algorithms are much more efficient (up to almost two orders of magnitude) on MSP430 than indicated by results in the literature.
Christian Franck, Johann Großschädl

Networking and Communications Technology for IoT

Frontmatter
Multi-layered Model for Performance Evaluation of oneM2M-Based IoT Solution
Abstract
In this paper we evaluate the impact of standards in terms of performance and their applicability in the field of IoT system design and deployment. We focus on the global IoT oneM2M standard. Our objective is to evaluate a oneM2M-based IoT solution regarding different relevant Key Performance Indicators. We propose a multi layered-model of an IoT standardized solution, able to tackle applicative, infrastructure and deployment aspects. Based on this model, we are able to globally evaluate and analyze, through simulation, the adequacy of a deployment with respect to the initial applicative constraints and the chosen oneM2M standard implementation. In our case, the constraints are mix-critical coming from the e-Health remote monitoring of patient by their physician but also the management of the patient in case of vital emergency situation. By tuning the system configuration and parameters of the proposed applicative scenario, we evaluate, by simulation, the KPIs of a oneM2M-based IoT solution by exploiting (1) the different features of the standard, (2) the capabilities of the underlying infrastructure, and (3) the performance of the oneM2M stacks used in the solution. The simulation and performance evaluation are based on two tools developed by the authors. One is a specific profiler for oneM2M open-source stack, whereas the simulation and performance evaluation is build on top of the OMNeT++ discrete event simulator.
Samir Medjhah, Thierry Monteil, Marie-Agnès Peraldi-Frati, Luigi Liquori
An Information-Theoretic Approach for Anomaly Detection in RPL-Based Internet of Things
Abstract
In recent years, cyber-attacks have increased significantly in both volume and sophistication, making the detection of security violations a crucial feature in computer systems. This is particularly true in the Internet of Things (IoT), where devices are vulnerable to failures and malicious attacks due to their resource-constrained nature. Given the proliferation of new security threats, anomaly-based detection approaches are essential for intrusion detection and prevention systems to effectively defend against attackers. This paper proposes an information-theoretic approach based on entropy to establish an anomaly detection model. A real case study in IoT networks based on Routing Over Low power and Lossy networks (RPL) illustrates the application of the proposed approach. Preliminary experimental results demonstrate that our method is both practical and extendable.
Vinh Hoa La, Edgardo Montes de Oca, Ana Cavalli
Formal Development of a Delay-Tolerant Multicast Protocol for Wireless Sensors
Abstract
We consider environmental monitoring in a remote area with limited connectivity where motes can join and leave the network arbitrarily, the topology is dynamic, transmission is highly unreliable, power is restricted, data points are sampled in large intervals, the data volume is low, a delay of the reception of data points can be tolerated, and motes have large memory. We propose a new protocol with blind multicasting of data points, blind multicasting of acknowledgements, and caching of data points and acknowledgements. This paper presents the protocol by stepwise refinement with Event-B. The unreliability of transmissions is modelled by finitary fairness. Rodin is used to prove the correctness and an upper bound for the transmission delay. The protocol has been implemented using LoRa for the physical layer.
Emil Sekerinski, Tianyu Zhou

Artificial Intelligence and Machine Learning Technologies for IoT

Frontmatter
Graph-Based Classification of IoT Malware Families Enhanced by Fuzzy Hashing
Abstract
The proliferation of Internet of Things (IoT) devices has led to an increase in IoT malware, posing a significant cybersecurity threat. Detecting and mitigating this threat is challenging due to the diverse CPU architectures in IoT malware families and the limited resources of IoT devices. Specialized detection methods are needed to identify malware across different platforms, while lightweight mechanisms are required to minimize resource strain. This paper introduces a novel graph-based framework, Aggregated Weighted Graph of Hashes (AWGH), to tackle the CPU diversity challenge. The framework leverages Function Call Graphs (FCGs) and fuzzy hashing to capture the structural and code characteristics of IoT malware. By utilizing static analysis techniques, the framework can efficiently group new malware samples and identify similarities with existing families, even in the case of unknown malware to mitigate potential risks before they cause significant damage. FCGs are generated using IDA Pro [1], and fuzzy hashes are calculated using ssdeep [2]. The framework is implemented in Python and evaluated using a dataset from VirusTotal [3] through 10-fold cross-validation. The experimental results demonstrate the effectiveness of the proposed framework in accurately classifying the IoT malware into IoT malware families across various CPU architectures (MIPS, ARM, i386, PowerPC, and AMD64).
Nastaran Mahmoudyar, Ali A. Ghorbani, Arash Habibi Lashkari
Error Resiliency and Adversarial Robustness in Convolutional Neural Networks: An Empirical Analysis
Abstract
The increasing pervasiveness of Artificial Intelligence (AI), and Convolutional Neural Networks (CNNs) in edge-computing and Internet of Things applications pose several challenges, including the hunger for computational and power resources of predictive models, and their robustness w.r.t. security threats, e.g., adversarial attacks. As for the former, the approximate computing emerged as one of the most promising solutions to lower the computational effort of AI, since the output of approximate application is usually barely distinguishable from the exact one. Nevertheless, alterations to predictive models through approximation may actually jeopardize inner characteristics of CNNs, such as their adversarial robustness, that is their ability to discern legitimate inputs from systematically crafted malicious ones.
In this paper, we investigate the vulnerability of the approximate CNNs to adversarial attacks. Specifically, we target approximate CNNs while resorting to different adversarial attacks in an aversion scenario, and we empirically prove approximation may actually compromise adversarial robustness.
Mario Barbareschi, Salvatore Barone, Valentina Casola, Salvatore Della Torca

Cyber Security/Privacy/Trust for IoT and CPS

Frontmatter
A Blockchain and IPFS-Enhanced Model for Attack Detection and Resource Efficiency
Abstract
The Social Internet of Things (SIoT) facilitates seamless interactions between IoT devices, providing users with quick and convenient services. However, this domain is vulnerable to manipulation by malicious nodes that issue false recommendations and services to inflate their reputation, leading to trust-related attacks. Developing trust models to detect these attacks in each interaction is challenging due to the complexity of the patterns and features required for accurate prediction. Furthermore, trust metrics are not consistently updated for each node, resulting in inefficiencies and unnecessary resource consumption. To address these challenges, we propose a system that analyzes the context of the current interaction and incorporates temporal factors to monitor node behavior. Our approach employs a decentralized system based on blockchain and IPFS storage, reducing costs and making the process of trust evaluation more efficient and practical for real-time scenarios. This method enhances the detection of trust-related attacks while optimizing resource allocation and execution time.
Raouf Jmal, Mariam Masmoudi, Ikram Amous, Florence Sèdes
Hardware Trojan Key-Corruption Detection with Automated Neural Architecture Search
Abstract
This work presents a model hardware trojan which intermittently is capable of corrupting an encryption operation occurring on a device. It asks whether this trojan can be detected via power-based, side-channel attacks only instrumenting the encryption itself, not the control flow of the trojan itself. By applying Automated Machine Learning techniques to search neural architecture, a classification of corrupted encryption operations is able to completely identify whether the operation corresponded with a corrupted operation or not. Through a number of experiments, we demonstrate this fact holds regardless of variable or constant plaintext, rotating encryption keys, or even with different corrupted keys.
Franco Mezzarapa, Jenna Goodrich, Andey Robins, Mike Borowczak

IoT or CPS Applications and Use Cases

Frontmatter
Actuation Conflict Management in Internet of Things Systems DevOps: A Discrete Event Modeling and Simulation Approach
Abstract
In IoT DevOps, simulating actuation conflict management is crucial for enhancing conflict detection and resolution in concurrent IoT applications. This paper introduces a new discrete event modeling and simulation approach for IoT systems during the design phase. Its objectives are to identify potential conflicts arising from competing smart applications accessing shared actuators or physical properties and to validate actuation conflict management specifications aimed at resolving conflicts among concurrent access attempts to IoT devices. The formalism of discrete event system specification is employed to model IoT systems formally, incorporating an actuation conflict management simulation model. This aids designers in the resolution of actuation conflicts.
Laurent Capocchi, Jean-Francois Santucci, Jean-Yves Tigli, Thibault Gomnin, Stephane Lavirotte, Gerald Rocher
Leveraging Task-Specific VAEs for Efficient Exemplar Generation in HAR
Abstract
The emerging technologies of smartphones and wearable devices have transformed Human Activity Recognition (HAR), offering a rich source of sensor data for building an automated system to recognize people’s daily activities. The sensor-based HAR data also enables Machine Learning (ML) algorithms to classify various activities, indicating a new era of intelligent systems for health monitoring and diagnostics. However, integrating ML into these systems faces the challenge of catastrophic forgetting, where models lose proficiency in previously learned activities when introduced to new ones by users. Continual Learning (CL) has emerged as a solution, enabling models to learn continuously from evolving data streams while reducing forgetting of past knowledge. Within CL methodologies, the use of generative models, such as Variational Autoencoders (VAEs), for example, has drawn significant interest for their capacity to generate synthetic data. This reduces storage demands by creating on-demand samples. However, the application of VAEs with a CL classifier has been limited to low-dimensional data or fine-grained features, leaving a gap in harnessing raw, high-dimensional sensor data for the HAR model. Our research aims to bridge this gap by constructing VAEs with a filtering mechanism for direct training with raw sensor data from the HAR dataset, enhancing CL models’ capability in class-incremental learning scenario. We demonstrate that VAE with a boundary box sampling and filtering process significantly outperforms both traditional and hybrid exemplar CL methods, offering a more balanced and diverse training set that enhances the knowledge acquisition of the model. Our findings also emphasize the importance of sampling strategies in the latent space of VAEs to maximize data diversity, crucial for recognizing the variability in human activities for better representation of each activity in each CL task.
Bonpagna Kann, Sandra Castellanos-Paez, Romain Rombourg, Philippe Lalanda
The Role of Ethics in Smart Homes – A Workshop-Based Approach
Abstract
Smart homes are increasingly popular and offer users multiple benefits, such as increased security, entertainment, health, and energy efficiency. But smart homes also raise ethical challenges. Analyzing ethical risks in smart homes requires an approach that can reveal and analyze the complex consequences of unethical IoT use. Such an analysis, however, is cumbersome and requires including many aspects and stakeholder perspectives. There is a lack of methods to analyze smart homes ethically and document such research results for continual evaluation over time as the smart home and our understanding of its ethics inevitably evolve and change. This work aims to design a workshop methodology to support systematic ethical analyses of smart homes. It builds on previous work considering smart homes as digital ecosystems to contextually examine ethical risks and challenges. A group of research participants were asked to undergo the workshop to evaluate its usefulness in supporting ethical discussions and documenting insights systematically. The results show the feasibility of the workshop design in conducting ethical analyses and eliciting system requirements for smart homes. Several unethical use cases are discussed, such as IoT gaslighting and surveillance concerns related to child users.
Sally Bagheri, Andreas Jacobsson
Digital Twin-Based Security Orchestration, Automation and Response for IoT and CPS
Abstract
The digitisation leveraging technologies in the Internet of Things (IoT) and Cyber-Physical Systems (CPS) has been largely adopted together with the Digital Twin (DT) paradigm. However, the distributed and heterogeneous nature of IoT or CPS poses significant challenges in safeguarding against diverse attack surfaces, including physical devices, network infrastructures, and third-party integration. Furthermore, the evolving security threats and potential cascading effects from cyber attacks add another layer of complexity to the security landscape. Therefore, in this paper, we propose a digital twin-based security orchestration automation and response framework, striving for the business continuity (SOAR4BC). Leveraging system contexts from the DT in combination with security intelligence from the security tools gives us a holistic context for SOAR, which has not been seen in the existing approaches. By subjecting tampered data and distributed denial of service (DDoS) detection to rigorous experimental evaluation, we substantiate the efficacy and reliability of the SOAR4BC framework in detecting and responding to security policy violations within simulated digital twin environments. This validation serves as a compelling proof of concept, highlighting the SOAR4BC framework’s robustness in addressing cyber threats. Our work offers novel insights into the convergence of digital twin technology and cybersecurity, illuminating the unique challenges and opportunities inherent in DT-based IoT and CPS systems.
Phu H. Nguyen, Ashish Rauniyar, Toni Valtteri Niemi
GreenMov: A Fiware Based Interoperable Solution to Reduce the Environmental Impact of Mobility
Abstract
The recent advancements in the Internet of Things (IoT) have facilitated the deployment of numerous applications that enhance urban intelligence by improving the monitoring and control of city assets, such as lighting systems, traffic signals, and public transportation. However, these applications are often tailored to the specific needs of individual cities, limiting their replicability in other locations primarily due to the lack of interoperability in communication signals between assets. Despite various initiatives to establish standards for interoperability, real-world implementations frequently fall short of achieving fully interoperable systems that can be universally replicated, largely due to the absence of comprehensive solutions and implementation examples. This paper presents a fully interoperable use case for Green Mobility solutions in a smart city, utilizing air quality, traffic, and noise intensity data to provide transport recommendations to end-users. The implementation employs the NGSI-LD standard, Fiware data storage tools, and developed artificial intelligence-based algorithms to predict the transport situation for the following day and offer relevant traffic recommendations. This work has resulted in the development of several data models and standardized forecast algorithms with accuracy exceeding 75% on the noise and traffic datasets at our disposal, thereby enabling the potential for replication in other locations.
Benoit Couraud, Mehdi Nafkha, Franck Dechavanne, Azeddine El Youssfi, Paulo Moura
Dynamic IoT Determination of Overall Heat Transfer Coefficient in a Portable Cabin in Kuwait
Abstract
Two portable cabins (size 2 m width, 2 m depth, 2.8 m height) were constructed using 75 mm Polyurethane sandwich panels (insulation walls) to simulate low energy buildings in Kuwait. Experimental measurements were conducted using eight K-type thermocouple probes on interior/exterior surfaces of the cabins’ walls to dynamically evaluate overall heat transfer coefficient during the hot months of August and September 2023 using IoT data storage. The collected data was further curated to ensure quality by treating negative values, handling potential division by zero issues, and filtering out large numbers. The dynamic average overall heat transfer coefficients were found to be 0.186, 0.198, 0.203, and 0.206 W/m2K for the East, North, West, and South facing walls, respectively. This is far better than the static (steady state) overall heat transfer coefficient of 0.2534 W/m2K, indicating that the dynamic behavior of the insulation walls is very promising, particularly by saving 21.76% cooling energy demands in summertime in Kuwait.
Ahmad Sedaghat, Mohammad Nazififard, Mohamad Iyad Al-Khiami
Backmatter
Metadaten
Titel
Internet of Things
herausgegeben von
Gaëtan Rey
Jean-Yves Tigli
Erwin Franquet
Copyright-Jahr
2025
Electronic ISBN
978-3-031-81900-1
Print ISBN
978-3-031-81899-8
DOI
https://doi.org/10.1007/978-3-031-81900-1