Cyber-Security of IoT in Post-Quantum World

The growth of quantum computing reveals a series of significant threats to cryptographic systems known today, such as RSA, ECC, and Diffie-Hellman, which are extremely important for Internet security, but are threatened by the increase in possible quantum attacks. The area of public and private key cryptography is vast. Therefore, this article addresses post-quantum cryptography (PQC) algorithms and the challenges of implementing them in Internet of Things (IoT) devices, given the difficulty of application due to restricted resources, such as RAM and clock speed . To this end, we will discuss the need for standardization of algorithms considering low-resource IoT devices, prioritizing RAM over ROM, meeting the needs of low-memory devices, which are lighter. The National Institute of Standards and Technology (NIST) has validated four cryptographic systems that are secure against quantum attacks. We analyze the implications of these systems on the infrastructure of current networks and will compare in detail these algorithms and their respective strengths, weaknesses, and trade-offs. We evaluate the feasibility of post-quantum encryption (PQC) algorithms in resource-constrained IoT devices using the ESP-WROOM-32 microcontroller. Three cryptographic schemes were implemented and compared.

In recent years, the Internet of Things (IoT) has attracted attention according to the development of sensing technology and wireless communication technology. IoT devices are interconnected to realize a specific service with sensors and actuation functions over several communication methods. Internet Protocol (IP) exists as a basic communication technology for realizing an IoT system. Since IPv4 is a mainstream protocol on the Internet, the lack of compatibility between IPv4 and IPv6 should be great issues to realize IoT services. Additionally, recent networks employ the Network Address Port Translation (NAPT) technology to protect a network and to reduce the number of required IPv4 global addresses. The introduction of NAPT causes an accessibility issue due to the blocking of incoming packets. As a new issue, the mobility of IoT devices also becomes a concern about seamless communication because a change of IP address causes a disconnection of transport layer sessions. As a technique for solving these problems, the authors have developed CYber PHysical Overlay Network over Internet Communication (CYPHONIC) that supports accessibility and seamless mobility over the Internet. Since high-performance IoT devices typically implement the full IP stack, both IPv4 and IPv6 can be supported. Therefore, CYPHONIC can provide inter-connectivity among IoT devices. On the contrary, resource-constrained IoT devices with a low power operation support only the limited IP stack such as IPv6 over Low power Wireless Personal Area (6LowPAN). This paper proposes a new interconnectivity mechanism between IP networks and 6LowPAN by extending CYPHONIC. Therefore, the proposed mechanism can provide interconnectivity among 6LowPAN devices and IP devices to collaborate on a service. The proposed system introduces a gateway device for connecting between IP networks and 6LowPAN. Additionally, it also extends the signaling process of CYPHONIC to support 6LowPAN based IoT devices. Since the object of this paper is confirming the feasibility of the proposed mechanism, we have designed the implementation with Thread protocol that support 6LowPAN.

Internet of things (IoT) has become a trend that attracts many scientists, businesses, and application research and development organizations in recent years. However, security for IoT devices in general and resource-constrained IoT devices, in particular, has become an important issue. This chapter will present an overview of resource-constrained IoT devices and explain the characteristics of resource-constrained IoT devices that make integrated information security and safety solutions currently 42unavailable, such as traditional electronic devices (i.e., PCs, laptops, and servers). This chapter will also go through the issues related to the traditional electronic device agents and the resource-constrained IoT device agents. Unlike traditional electronic devices with mostly i386-based processors and Windows operating systems, resource-constrained IoT devices use a variety of different processors such as MIPS32, MIPS64, ARM, ARM64, PowerPC, and SPARC and need limited resources (i.e., storage and processing). The data collected by the agent on the resource-constrained IoT devices are of different varieties (i.e., from network traffic data (PCAP) to system data of the device (syscalls, PID, etc.)); therefore, detecting and preventing cyberattacks against resource-constrained IoT devices is effective and practical when preventing cyberattacks is handled directly on resource-constrained IoT devices.

The rapid proliferation of Internet of Things (IoT) devices, coupled with their inherent security vulnerabilities, has significantly expanded the attack surface, intensifying threats such as man-in-the-middle attacks, traffic hijacking, and distributed denial-of-service (DDoS) attacks, thereby posing serious risks to the security and reliability of the entire ecosystem. The network traffic associated with IoT devices is diverse and dynamic, often exhibiting complex structural features such as periodic fluctuations, varying packet sizes, and time-varying patterns that make detection challenging. Although deep learning has demonstrated strong capabilities in efficiently identifying complex and dynamic malicious traffic through powerful feature extraction and adaptive learning abilities, its high model complexity, substantial computational demands, and large parameter sizes hinder direct deployment on resource-constrained IoT devices. In order to tackle this issue, this paper proposes a malicious traffic detection framework for the Internet of Things (IoT) based on multi-teacher knowledge distillation. The proposed model, termed the Lightweight Multi-Teacher Spatiotemporal Distillation Neural Network (LMT-SDNN), employs two high-performance teacher models: Residual Inception and a One-Dimension Convolution Netural Network (1D-CNN) integrated with idirectional Long Short-Term Memory (BiLSTM), to effectively capture the complex structural features of network traffic. Furthermore, a novel Time-Related Window Loss (TRW) function is design to enhance the student’s ability to capture temporal features, thereby improving its overall performance. The effectiveness of LMT-SDNN is validated through comparisons with five baseline models on two publicly available datasets, ToN_IoT and BoT_IoT. Experimental results show that LMT-SDNN achieves a compression rate of over 99% in both model complexity and parameter count, while maintaining an accuracy exceeding 99%, indicating its strong potential for multiclass malicious traffic detection in IoT environments.

The Internet of Things (IoT) now underpins critical systems like healthcare monitors, smart home, smart grids, automotive V2V & V2X networks, and industrial controls but it still relies on classical cryptography, whose security assumptions are vulnerable to quantum computers. A large quantum computer could run Shor’s algorithm to break RSA/ECC and use Grover’s algorithm to accelerate brute-forcing of symmetric keys, undermining the confidentiality and authentication of data in these IoT applications. Researchers have proposed quantum cryptographic solutions. Quantum key distribution (QKD) and quantum random number generators (QRNGs) exploit quantum mechanics to detect eavesdropping and deliver provably secure key material. For example, chip-scale QRNGs can be embedded in IoT sensors to generate true random keys, boosting resistance to brute-force attacks. These quantum techniques provide information-theoretic secrecy, for instance, QKD-generated keys have been used to authenticate smart-grid SCADA. However, most IoT devices are resource-constrained, making direct deployment of conventional QKD hardware challenging. Post-quantum cryptography (PQC) complements quantum methods by providing algorithms such as lattice-based, hash-based that resist quantum attacks on existing hardware. Standardized PQC can be deployed via software updates on legacy IoT nodes, enabling scalable and quantum-safe security within current constraints. Researchers are exploring lightweight PQC for connected cars and e-health sensors. Hybrid approaches combining QKD/QRNG with PQC offer a layered defense. Since QKD protocols are regarded as ‘future-proof’ and PQC standards mature, their combined use will be essential for robust, quantum-safe IoT security. Miniaturizing quantum components such as integrated QKD/QRNG modules and optimizing lightweight PQC schemes for low-power IoT devices are future research directions to secure IoT networks.

The explosive proliferation of Internet of Things (IoT) devices is generating an incomprehensible amount of data. Machine learning plays an imperative role in aggregating this data and extracting valuable information for improving operational and decision-making processes. In particular, emerging machine intelligence platforms that host pre-trained machine learning models are opening up new opportunities for IoT industries. While those platforms facilitate customers to analyze IoT data and deliver faster and accurate insights, end users and machine learning service providers (MLSPs) have raised concerns regarding security and privacy of IoT data as well as the pre-trained machine learning models for certain applications such as healthcare, smart energy, etc. In this paper, we propose a cloud-assisted, privacypreserving machine learning classification scheme over encrypted data for IoT devices. Our scheme is based on a three-party model coupled with a two-stage decryption Paillier-based cryptosystem, which allows a cloud server to interact with MLSPs on behalf of the resource-constrained IoT devices in a privacy-preserving manner, and shift load of computation-intensive classification operations from them. The detailed security analysis and the extensive simulations with different key lengths and number of features and classes demonstrate that our scheme can effectively reduce the overhead for IoT devices in machine learning classification applications.

This study proposes a health monitoring system for snoring detection utilizing Tiny Machine Learning (TinyML) models, specifically designed for resource-constrained wearable Internet of Things (IoT) devices. This research addresses significant constraints associated with running machine learning models on IoT devices, such as latency, limited memory, and low computational resources. These parameters are essential for real-time monitoring in healthcare applications, where prompt response is critical. The research focuses on developing a TinyML model capable of identifying specific audio patterns related to snoring during sleep. Experimental evaluations conducted in real-world sleep environments with the TinyML model deployed on resource-constrained wearable IoT devices. The evaluation results show that the proposed model achieves high accuracy while utilizing minimal computational resources and without introducing latency issues. Through the integration of audio (Syntiant) and advanced audio preprocessing techniques, the proposed system improves the efficiency of the TinyML model on wearable devices. The quantized TinyML model achieved an accuracy of 95.85% with a low latency of 48 ms, utilizing only 17.0K of RAM and 34.07K of flash memory for real-time snoring classification. This study highlights the benefits of practical deployment of the TinyML model for snoring detection on resource-constrained wearable IoT devices, demonstrating that such models can operate effectively within the constraints of current wearable technology.

Internet of Things (IoT) has revolutionized the world with its innovative and precise solutions catering a broad spectrum of applications. Smart cities, smart financial services, smart logistics, smart agriculture systems, smart healthcare management system are few applications worth mentioning. Device authentication and data security is a major concern in IoT networks as the devices used in these networks have inadequate computational processing and memory storage resources. Privacy is of paramount importance in applications such as healthcare and finance. As complex algorithms cannot be implemented on the resource-constrained IoT devices, research in the aforementioned areas is still an active subject. In this paper, we propose a mutual device and server authentication protocol using Physical Unclonable Functions (PUFs) for IoT devices. PUFs can be used to uniquely identify the devices and can also be used to generate random keys. As the name suggests, the keys generated by the PUFs cannot be reproduced. Using this unclonability feature, an authentication protocol has been developed. An IoT based temperature and humidity sensing module is developed to integrate the protocol and validate its effectiveness in providing the device authentication and data security.

In the rapidly developing world of the Internet of Things (IoT), data security has become increasingly important since massive personal data are collected. IoT devices have resource constraints, which makes traditional cryptographic algorithms ineffective for securing IoT devices. To overcome resource limitations, lightweight cryptographic algorithms are needed. To identify research trends and patterns in IoT security, it is crucial to analyze existing works, keywords, authors, journals, and citations. We conducted a bibliometric analysis using performance mapping, science mapping, and enrichment techniques to collect the necessary information. Our analysis included 979 Scopus articles, 214 WOS articles, and 144 IEEE Xplore articles published during 2015–2023, and duplicates were removed. We analyzed and visualized the bibliometric data using R version 4.3.1, VOSviewer version 1.6.19, and the bibliometrix library. We discovered that India is the leading country for this type of research. Archarya and Bansod are the most relevant authors; lightweight cryptography and cryptography are the most relevant terms; and IEEE Access is the most significant journal. Research on lightweight cryptographic algorithms for IoT devices (Raspberry Pi) has been identified as an important area for future research.

The proliferation of the Internet of Things (IoT) has brought unprecedented connectivity, introducing complex trust and reputation management challenges across vast, heterogeneous networks. This paper introduces the DeepTrust framework, a novel approach leveraging deep learning algorithms to assess and manage trust and reputation in IoT environments dynamically. We demonstrate the framework’s superiority in accurately identifying trustworthy and untrustworthy devices through extensive experiments, significantly enhancing IoT security. DeepTrust has demonstrated marked superiority over existing methods, showcasing enhanced accuracy in identifying trustworthy versus untrustworthy devices, thereby significantly bolstering IoT network security. Specifically, our results reveal an improvement in accuracy by 15%, precision by 20%, and recall rates by 18% compared to conventional models, highlighting DeepTrust’s effectiveness in real-time, adaptive trust assessments. There are several avenues for enhancing and expanding the DeepTrust framework. Future research will explore optimization techniques for reducing computational demands, enabling deployment on resource-constrained IoT devices. Additionally, incorporating incremental learning mechanisms could improve the framework’s adaptability to new and changing IoT environments. Enhancing data privacy and security measures within the framework constitutes another critical development area, ensuring the protection of sensitive information used in trust assessments. Lastly, extending the framework’s applicability across various IoT domains and applications presents a promising direction, aiming to establish a universal trust management solution adaptable to the unique requirements of different IoT ecosystems. By outlining these potential future directions, we aim to highlight the current achievements of the DeepTrust framework and chart a course for its continued development and refinement. This comprehensive approach underscores our commitment to advancing the field of IoT trust and reputation management, paving the way for more secure and reliable IoT networks.

INTRODUCTION: In contemporary society, everyday operations are greatly improved by the Internet of Things (IoT), which connects physical devices to provide digital services. IoT technology offers unified services and streamlines activities across various domains, ranging from remote monitoring to sophisticated welfare systems. However, the growing number of IoT devices presents a security concern. Many of these devices are susceptible to exploitation, leading to diverse vulnerabilities. OBJECTIVES: Resource-constrained IoT devices become prime targets for botnet attacks, manifesting in various forms and penetration methods. Despite numerous research efforts introducing multiple approaches for detecting botnet attacks in IoT, existing methods often fail to achieve satisfactory detection rates. METHODS: Additionally, these approaches struggle to comprehensively analyze the diverse communication networks within the expansive realm of IoT devices. This study proposes an innovative machine-learning framework for detecting IoT botnet threats to address these limitations. RESULTS: This conceptual framework exhibits a remarkable capability to identify a spectrum of botnet attacks, showcasing a detection accuracy of 99.5 per cent, significantly surpassing the performance of other prevalent machine-learning approaches. CONCLUSION: Through this research, we aim to enhance the security paradigm of IoT networks, ensuring robust protection against evolving botnet threats in the dynamic landscape of interconnected devices.

Deploying high-performance vision transformer (ViT) models on ubiquitous Internet of Things (IoT) devices to provide high-quality vision services will revolutionize the way we live, work, and interact with the world. Due to the contradiction between the limited resources of IoT devices and resource-intensive ViT models, the use of cloud servers to assist ViT model training has become mainstream. However, due to the larger number of parameters and floating-point operations (FLOPs) of the existing ViT models, the model parameters transmitted by cloud servers are large and difficult to run on resource-constrained IoT devices. To this end, this article proposes a transmission-friendly ViT model, TFormer, for deployment on resource-constrained IoT devices with the assistance of a cloud server. The high performance and small number of model parameters and FLOPs of TFormer are attributed to the proposed hybrid layer and the proposed partially connected feed-forward network (PCS-FFN). The hybrid layer consists of nonlearnable modules and a pointwise convolution, which can obtain multitype and multiscale features with only a few parameters and FLOPs to improve the TFormer performance. The PCS-FFN adopts group convolution to reduce the number of parameters. The key idea of this article is to propose TFormer with few model parameters and FLOPs to facilitate applications running on resource-constrained IoT devices to benefit from the high performance of the ViT models. Experimental results on the ImageNet-1K, MS COCO, and ADE20K datasets for image classification, object detection, and semantic segmentation tasks demonstrate that the proposed model outperforms other state-of-the-art models. Specifically, TFormer-S achieves 5% higher accuracy on ImageNet-1K than ResNet18 with 1.4× fewer parameters and FLOPs.

In the era of ubiquitous intelligence, the Internet of Things (IoT) holds the promise as a breakthrough technology to enable diverse applications that benefit societal problems. Yet interconnecting myriad heterogeneous IoT devices across various application domains remain a security challenge. Decentralized technology has recently emerged as a powerful primitive in building distributed applications to facilitate secure transactions between mutually distrustful parties in a trustworthy manner. Unfortunately, these decentralized protocols demand computing resources and power far beyond the reach of resource-constrained IoT devices, preventing the full adoption of distributed consensus platform in the IoT setting. In this article, we address the key bottleneck to enable blockchain in resource-constrained IoT devices. We propose a lightweight implementation of proof-of-work (PoW) mining with reconfigurable hardware primitives. By replacing the hash and cryptographic functions in classic blockchain protocol with secure and efficient hardware implementations, our proposed solution can significantly reduce hardware resources and power overheads of PoW mining, while improving the transaction speed of large-scale IoT systems. Finally, we demonstrate the algorithm by proposing an antispoofing solution for GPS navigation among lightweight IoT devices. As a replacement for position computation, a mining process generates the expected coordinates with the correct initial value and function configuration.

Using low-rank adaptation to fine-tune pre-trained neural network models has attracted widespread attention due to its advantages of low resource requirements, high precision, and no additional inference delay. However, most existing methods are designed for large language models based on the Transformer structure and lack adaptation to convolutional neural networks (CNN) widely used on Internet of Things (IoT) devices. In addition, IoT devices are usually deployed outdoors and collect large amounts of data that are affected by environmental conditions. Providing a highly robust CNN model is a prerequisite for providing high-quality services. To this end, this paper proposes a highly robust seed-filter adaptation method (SeFA) for pre-trained CNNs. SeFA introduces an adaptation branch with the same structure as the backbone network. In the adaptation branch, some filters are first designated seed filters and grouped. Then, additional filters are generated from the grouped seed filters and nonlinear transformation functions (NLFs) with different hyperparameters. The seed filters' parameters are updated during model training, and the NLF hyperparameters are randomly initialized and frozen. Both grouping seed filters and configuring NLFs with non-learnable hyperparameters can improve the model's robustness. This is because grouping seed filters can generate diverse filters on demand without increasing the model's complexity, and the NLFs' rules can regularize the model. It is worth mentioning that the number of fine-tuning parameters of the pre-trained model that can adapt to downstream tasks can be flexibly controlled by specifying the number of seed filters. The key idea of this paper is to propose SeFA with flexible, controllable, learnable parameters, high robustness, and adaptability to pretrained CNN models, thereby facilitating fine-tuning on resource-constrained IoT devices and providing highly robust visual services. Experimental results on the CIFAR-10, CIFAR-10-C, CIFAR-100, CIFAR-100-C, and Icons-50 datasets demonstrate that the proposed SeFA outperforms other state-of-the-art methods. Specifically, on the ResNet-152 model and the CIFAR-10-C dataset, the accuracy of our SeFA is about 7% higher than that of the full fine-tuning method. In addition, we verify the efficiency of SeFA-based models for fine-tuning and inference on the device.

With the proliferation of Internet of Things (IoT) devices, cybersecurity has become an increasingly pressing concern. Due to their inherent vulnerabilities and the severity of the threats posed by malware, robust malware detection strategies for IoT devices are in high demand. However, the constraints imposed by the limited memory and processing power of IoT devices, alongside with the varied nature of these devices, have limited the development of effective malware detection strategies. Existing solutions primarily use either local computation or cloud-based services, each with its own set of challenges. In this paper, we propose a hybrid malware detection method that leverages both local and cloud-based computation to address these challenges. This method, based on similarity computations, optimizes the trade-off between computational resources and latency, offering an effective approach for malware detection in IoT devices. Through experimental evaluation using a substantial IoT malware dataset, we demonstrate that our proposed approach achieves an optimal balance between local and cloud-based computations, providing a promising solution for malware detection in IoT context.