Computer Networks and Communications

Security Issues in IoT: Perspective Review

2025-02-18T14:48:49+08:00

The Internet of Things (IoT) has revolutionized various sectors, including healthcare, wearables, automotive, smart cities, agriculture, manufacturing, business, and home automation. As technology continues to shape everyday life, ensuring the security of IoT systems has become a critical concern. IoT devices, equipped with multiple sensors and processors, collect vast amounts of data, making them vulnerable to cyber threats. Like other connected systems, these devices are susceptible to attacks such as credential theft, firmware exploits, and hardware-based intrusions. IoT security involves safeguarding physical components, data, and network connections against unauthorized access and malicious activities. Key security measures include software and firmware updates, credential protection, device authentication, encryption, disabling unnecessary IoT functionalities, and Domain Name System (DNS) filtering. However, securing IoT networks is challenging due to the diverse range of devices, lack of standardized security protocols, and the widespread use of open-source software. This article explores security vulnerabilities across different IoT layers, examines existing mitigation strategies, and discusses potential research directions for enhancing IoT security.

A Comparative Study of Wi-Fi Technologies in Wireless Sensor Networks

2025-01-16T09:18:49+08:00

Wireless Sensor Networks (WSNs) have become a fundamental part in various Internet of Things (IoT) applications, such as smart cities, industrial automation, and environmental monitoring. This paper focuses on utilizing Wi-Fi technology within WSNs due to its high data rates and widespread infrastructure, which are essential for real-time monitoring and control applications. We conduct a comparative analysis of popular wireless communication technologies, including ZigBee, Bluetooth, and LoRa, and identify Wi-Fi as a suitable option for scenarios requiring extensive data transmission. The system design involves using ESP32 microcontrollers as sensor nodes to collect and transmit data wirelessly to a central gateway. Performance evaluation demonstrates the reliability and efficiency of the Wi-Fi-based WSN, with notable improvements in data transmission reliability, reduced power consumption using Wi-Fi 6's Target Wake Time (TWT) feature, and low-latency performance suitable for real-time applications. Despite the challenges posed by high power consumption and susceptibility to interference, hybrid solutions combining Wi-Fi with other low-power technologies like ZigBee or LoRa are suggested to enhance energy efficiency and coverage. This study highlights Wi-Fi's potential in WSNs and its applicability to a wide range of IoT implementations.

Deep Learning Using Path Length Prediction for Internet of Things

2025-02-14T14:30:30+08:00

Sensors are employed in the Internet of Things (IoT) to collect data and establish connections with the internet. An instance of IoT can be seen in a tree-like topology constructed using wireless links. When a topology graph has a path from its root node to any other leaf or child node, and this path is influenced by the quality of wireless connections, it is known as a Destination Oriented Directed Acyclic Graph. The root node of the tree topology is responsible for implementing source routing for downstream paths to the leaf nodes of the tree. If the longest path for any node in a tested network graph, including the root, is determined by the maximum hop count, then the graph is considered to be connected. The real world and its applications are impacted by issues related to network connectivity in IoT. Models are employed to examine how changes in link probability and hop count affect the connectivity of the graph. In this research, the proposed Deep Learning (DL) model is evaluated using the Keras regression model. The simulated dataset is generated using the Cooja emulator. The link probability serves as a feature to predict the maximum hop count in the IoT. The predicted hop count based on the link probability aligns accurately with the tested data.

Exploring Signal Interpolation: A Comparative Study of Convolution, Regression, and SVM Methods

2025-03-18T15:12:52+08:00

Signal interpolation plays a critical role in various signal processing applications, including wireless communications, image processing, and radar systems. Accurately reconstructing signals from decimated samples is essential for maintaining data integrity and improving transmission efficiency. To this extent, this paper presents a comparative study of five signal interpolation methods: convolution with three types of deterministic signals (triangular, rectangular and sinc signal), statistical linear regression and support vector machine (SVM). All these methods were applied on a sinusoidal signal corrupted by noise at different signal to noise ratio (SNR) values and on a QPSK (Quadrature Phase Shift Keying) modulated signal with 25 different decimation factors. The comparison between the investigated methods was made based on the inter-correlation coefficient, Euclidean distance and determinism for sinusoidal signal corrupted by noise. Two additional parameters, namely Euclidean letter distance and Bit Error Rate (BER), were defined and used for the QPSK modulated signal. Our findings suggest that for the sinusoidal signal corrupted by noise convolution with sinc function outperforms the other methods in terms of Euclidian distance in at least 98.57% of the cases and at least 95.71% of the cases in terms of inter-correlation coefficient. In the case of QPSK modulated signal it is the SVM method which surpasses all the other methods in terms of intercorrelation coefficient and Euclidean distance, in 80% and 88% of the cases respectively. If the Euclidean letter distance and the Bit Error Rate are considered for comparison, in the case of the QPSK modulated signal, convolution with sinc function was found to outperform the other investigated methods for at least 80% and 60% of the decimation factors respectively.

Estimating Throughput in Optical Backbone Networks Using Deep Neural Networks

2024-12-19T10:23:47+08:00

Optical backbone networks, which use optical fibres as the transmission medium, form the core infrastructure used by network operators to deliver services to users, as well as by Internet companies to route traffic between data centres. The network throughput is a key parameter in the analysis of the networks' performance. However, its determination can be a complex process that involves long computation times, since aspects related to both the physical and network layers need to be accounted for. To face this challenge, we propose a machine learning solution: a deep neural network (DNN) model, that has the goal of estimating the values of the network throughput and of a closely related parameter, average channel capacity, accurately and with short computation times. The simulation results indicate that the DNN model accurately predicts both outputs, with mean relative errors of 6.17% for the network throughput and 2.84% for the average channel capacity. These predictions are made in just a few milliseconds, providing a significant advantage over the heuristic routing algorithms, which can take up to tens of seconds in larger networks.

Mitigating Default Password Risks in CCTV: A Qualitative Study to Guide Recommendations for Device Makers

2024-12-06T10:22:55+08:00

The rapid growth of Internet of Things (IoT) devices has brought unmatched convenience, connectivity, and significant cybersecurity issues. One of IoT devices' predominant security risks is default passwords, making them vulnerable to various attacks and exploits. CCTV is highly susceptible to security breaches due to the reliance on default passwords. This paper identifies the risks associated with default passwords in CCTV and explores how they can be mitigated. Qualitative research was conducted to achieve this. Qualitative data was gathered through interviews with security experts, manufacturers, and CCTV end-users, and thematic analysis was subsequently analyzed. Through the research, the authors identified common security vulnerabilities and risks linked to default passwords in CCTV and employed password policies and authentication protocols. They recommended best practices to mitigate these risks. The results of this study have significant consequences for the area of IoT security, offering a broad understanding of the risks linked to default passwords in IoT devices, identifying optimal practices for mitigating them, and contributing valuable observations to more comprehensive discussions on IoT security. Eventually, the overarching goal of this study is to increase the safety and privacy of both individuals' and organizations' IoT devices and promote liable and ethical use of this technology.

A Novel Approach to Enhance the Security and Efficiency of Binary Ring-LWE for IoT Resource-Constrained

2024-11-26T09:42:16+08:00

The rapid expansion of the Internet of Things (IoT) brings a vast proliferation of network connections. This surge in connectivity significantly increases the risk of private data exposure during transmission and processing. Traditional public key encryption schemes face considerable challenges due to their high computational complexity and vulnerability to quantum attacks. Recently, Lattice-based cryptography, particularly the Binary Ring Learning With Errors (BRLWE) paradigm, has garnered significant attention for its quantum resistance and lightweight computational requirements. However, BRLWE remains vulnerable to physical attacks, especially Side-Channel Attacks (SCA). This paper proposes a novel 3-Decomposition Karatsuba multiplication-based random shuffling scheme to enhance both the efficiency and security of BRLWE. We evaluate the security performance of our proposed scheme against quantum hybrid attacks and SCAs. We assess the performances of different Karatsuba multiplication techniques in terms of computation cost, energy consumption and memory usage to make choose which Karatsuba technique is suitable for our proposal. Our experimental results show that our proposed approach provides the lowest encryption computation time of 18.97 ms and decryption computation time of 9.53 ms compared to the BRLWE and its improved versions. Furthermore, it improves the security level while it decreases the computation time of the original BRLWE by 32.49% and 20.58%, for the encryption and decryption phases, respectively.

Post-Quantum Key Agreement Protocols Based on Modified Matrix-Power Functions over Singular Random Integer Matrix Semirings

2024-12-06T14:34:48+08:00

Post-quantum cryptography is essential for securing digital communications against threats posed by quantum computers. Researchers have focused on developing algorithms that can withstand attacks from both classical and quantum computers, thereby ensuring the security of data transmissions over public networks. A critical component of this security is the key agreement protocol, which allows two parties to establish a shared secret key over an insecure channel. This paper introduces two novel post-quantum key agreement protocols that can be easily implemented on standard computers using rectangular or rank-deficient matrices, exploiting the generalizations of the matrix power function, which is a generator of NP-hard problems. We provide basic concepts and proofs, pseudocodes, and examples, along with a discussion of complexity.