The Internet of Medical Things (IoMT) has transformed healthcare by connecting medical devices, sensors, and patients, significantly improving patient care. However, the sensitive data exchanged through IoMT is vulnerable to security attacks, raising serious privacy concerns. Traditional key sharing mechanisms are susceptible to compromise, posing risks to data integrity. This paper proposes a Timestamp-based Secret Key Generation (T-SKG) scheme for resource-constrained devices, generating a secret key at the patient\'s device and regenerating it at the doctor\'s device, thus eliminating direct key sharing and minimizing key compromise risks. Simulation results using MATLAB and Java demonstrate the T-SKG scheme\'s resilience against guessing, birthday, and brute force attacks. Specifically, there is only a 9 % chance of key compromise in a guessing attack if the attacker knows the key sequence pattern, while the scheme remains secure against brute force and birthday attacks within a specified timeframe. The T-SKG scheme is integrated into a healthcare framework to securely transmit health vitals collected using the MySignals sensor kit. For confidentiality, the Data Encryption Standard (DES) with various Cipher Block modes (ECB, CBC, CTR) is employed.

Implicit poisoning in federated learning is a significant threat, with malicious nodes subtly altering gradient parameters each round, making detection difficult. This study investigates this problem, revealing that temporal analysis alone struggles to identify such covert attacks, which can bypass online methods like cosine similarity and clustering. Common detection methods rely on offline analysis, resulting in delayed responses. However, recalculating gradient updates reveals distinct characteristics of malicious clients. Based on this finding, we designed a privacy-preserving detection algorithm using trajectory anomaly detection. Singular values of matrices are used as features, and an improved Isolation Forest algorithm processes these to detect malicious behavior. Experiments on MNIST, FashionMNIST, and CIFAR-10 datasets show our method achieves 94.3% detection accuracy and a false positive rate below 1.2%, indicating its high accuracy and effectiveness in detecting implicit model poisoning attacks.

This paper presents a new edge detection process implemented in an embedded IoT device called Bee Smart Detection node to detect catastrophic apiary events. Such events include swarming, queen loss, and the detection of Colony Collapse Disorder (CCD) conditions. Two deep learning sub-processes are used for this purpose. The first uses a fuzzy multi-layered neural network of variable depths called fuzzy-stranded-NN to detect CCD conditions based on temperature and humidity measurements inside the beehive. The second utilizes a deep learning CNN model to detect swarming and queen loss cases based on sound recordings. The proposed processes have been implemented into autonomous Bee Smart Detection IoT devices that transmit their measurements and the detection results to the cloud over Wi-Fi. The BeeSD devices have been tested for easy-to-use functionality, autonomous operation, deep learning model inference accuracy, and inference execution speeds. The author presents the experimental results of the fuzzy-stranded-NN model for detecting critical conditions and deep learning CNN models for detecting swarming and queen loss. From the presented experimental results, the stranded-NN achieved accuracy results up to 95%, while the ResNet-50 model presented accuracy results up to 99% for detecting swarming or queen loss events. The ResNet-18 model is also the fastest inference speed replacement of the ResNet-50 model, achieving up to 93% accuracy results. Finally, cross-comparison of the deep learning models with machine learning ones shows that deep learning models can provide at least 3-5% better accuracy results.

The Internet of Things (IoT) is a promising technology for sensing and monitoring the environment to reduce disaster impact. Energy is one of the major concerns for IoT devices, as sensors used in IoT devices are battery-operated. Thus, it is important to reduce energy consumption, especially during data transmission in disaster-prone situations. Clustering-based communication helps reduce a node\'s energy decay during data transmission and enhances network lifetime. Many hybrid combination algorithms have been proposed for clustering and routing protocols to improve network lifetime in disaster scenarios. However, the performance of these protocols varies widely based on the underlying network configuration and the optimisation parameters considered. In this research, we used the clustering parameters most relevant to disaster scenarios, such as the node\'s residual energy, distance to sink, and network coverage. We then proposed the bio-inspired hybrid BOA-PSO algorithm, where the Butterfly Optimisation Algorithm (BOA) is used for clustering and Particle Swarm Optimisation (PSO) is used for the routing protocol. The performance of the proposed algorithm was compared with that of various benchmark protocols: LEACH, DEEC, PSO, PSO-GA, and PSO-HAS. Residual energy, network throughput, and network lifetime were considered performance metrics. The simulation results demonstrate that the proposed algorithm effectively conserves residual energy, achieving more than a 17% improvement for short-range scenarios and a 10% improvement for long-range scenarios. In terms of throughput, the proposed method delivers a 60% performance enhancement compared to LEACH, a 53% enhancement compared to DEEC, and a 37% enhancement compared to PSO. Additionally, the proposed method results in a 60% reduction in packet drops compared to LEACH and DEEC, and a 30% reduction compared to PSO. It increases network lifetime by 10-20% compared to the benchmark algorithms.

Orthogonal time frequency space (OTFS) modulation has recently found its place in the literature as a much more effective waveform in time-varying channels. It is anticipated that OTFS will be widely used in the communications of smart vehicles, especially those considered within the scope of Internet of Things (IoT). There are efforts to obtain customized traditional point-to-point single-input single-output (SISO)-OTFS studies in the literature, but their BER performance seems a bit low. It is possible to use cooperative communications in order improve BER performance, but it is noticeable that there are very few OTFS studies in the area of cooperative communications. In this study, to the best of the authors\' knowledge, it is addressed for the first time in the literature that better performance is achieved for the OTFS waveform transmission in a selective decode-and-forward (SDF) cooperative communication scenario. In this context, by establishing a cooperative communication model consisting of a base station/source, a traffic sign/relay and a smart vehicle/destination moving at a constant speed, an end-to-end BER expression is derived. SNR-BER analysis is performed with this SDF-OTFS scheme and it is shown that a superior BER performance is achieved compared to the traditional point-to-point single-input single-output (SISO)-OTFS structure.

In the era of ubiquitous computing, the challenges imposed by the increasing demand for real-time data processing, security, and energy efficiency call for innovative solutions. The emergence of fog computing has provided a promising paradigm to address these challenges by bringing computational resources closer to data sources. Despite its advantages, the fog computing characteristics pose challenges in heterogeneous environments in terms of resource allocation and management, provisioning, security, and connectivity, among others. This paper introduces COGNIFOG, a novel cognitive fog framework currently under development, which was designed to leverage intelligent, decentralized decision-making processes, machine learning algorithms, and distributed computing principles to enable the autonomous operation, adaptability, and scalability across the IoT-edge-cloud continuum. By integrating cognitive capabilities, COGNIFOG is expected to increase the efficiency and reliability of next-generation computing environments, potentially providing a seamless bridge between the physical and digital worlds. Preliminary experimental results with a limited set of connectivity-related COGNIFOG building blocks show promising improvements in network resource utilization in a real-world-based IoT scenario. Overall, this work paves the way for further developments on the framework, which are aimed at making it more intelligent, resilient, and aligned with the ever-evolving demands of next-generation computing environments.

Machine learning (ML) represents one of the main pillars of the current digital era, specifically in modern real-world applications. The Internet of Things (IoT) technology is foundational in developing advanced intelligent systems. The convergence of ML and IoT drives significant advancements across various domains, such as making IoT-based security systems smarter and more efficient. However, ML-based IoT systems are vulnerable to lurking attacks during the training and testing phases. An adversarial attack aims to corrupt the ML model\'s functionality by introducing perturbed inputs. Consequently, it can pose significant risks leading to devices\' malfunction, services\' interruption, and personal data misuse. This article examines the severity of adversarial attacks and accentuates the importance of designing secure and robust ML models in the IoT context. A comprehensive classification of adversarial machine learning (AML) is provided. Moreover, a systematic literature review of the latest research trends (from 2020 to 2024) of the intersection of AML and IoT-based security systems is presented. The results revealed the availability of various AML attack techniques, where the Fast Gradient Signed Method (FGSM) is the most employed. Several studies recommend the adversarial training technique to defend against such attacks. Finally, potential open issues and main research directions are highlighted for future consideration and enhancement.

Human skin has several receptors collaborating with the brain to provide appropriate \"decisions\" when applying stimuli. Several research articles state that biomimetic electronic skin (e-skin) is reportedly used for sensor-related applications and performs similarly to natural skin. However, research reporting the capability of the e-skin to make decisions and therefore react upon exposure to adverse conditions is still in its nascent stage. Herein, we report the development of an e-skin, ThermoSense, that can thermoregulate by making appropriate decisions. Thermoplastic polyurethane and multiwalled carbon nanotubes were used as the model composite. The heating and sensing capabilities of the optimized e-skin were studied in detail. In the study window, the e-skin demonstrated excellent electrothermal conversion efficiency by generating a temperature of 192 °C, consuming a power of 2.23 W. A finite element modeling (FEM) was adopted to determine the distribution of the filler in the case of the optimized e-skin and thus was used to probe the reason for the heating across the e-skin via mapping of the internal energy across the sample. FEM results and experimental findings are in strong agreement. Additionally, the e-skin demonstrated its capability to act as a thermal sensor with a 0.947% °C-1 sensitivity. To integrate the decision-making capabilities of the e-skin, an Internet of Things (IoT) brain console was made using the e-skin and electronic chips by leveraging More than Moore\'s concept. The IoT brain was automated with decision-making programming that was controllable via an in-house-developed mobile application. The console worked exclusively under simulated conditions. When there was a shift from the set point temperature, it started to heat. Postusage, the e-skin matrix was recycled, and the recycled e-skin demonstrated a marginal decrement in performance attributes. This study opens new avenues for developing decision-making e-skins for next-generation human-machine interphases.

Sleep quality is a critical factor in human health and well-being, with implications for various physiological and psychological processes. Traditional methods of sleep data collection are often limited by the quality and reliability of the data due to issues such as recall bias and subjective interpretation. This research aims to propose a novel framework that objectively measures and evaluates sleep quality using smart thermostats equipped with motion sensors, providing noninvasive and effortless sleep monitoring. The study conducts a comprehensive analysis of sleep patterns, exploring the relationship between activity sensors and sleep quality. By analyzing behavioral characteristics, the study identifies periods or clusters of days that require attention in terms of health and stress levels. The approach ensures privacy, ease of access, and integrates environmental factors, enabling a comprehensive understanding of an individual\'s sleep health. The findings suggest that this zero-effort technology can significantly enhance sleep monitoring at both individual and population levels, with implications for health monitoring, stress management, and personalized healthcare interventions. Future work will focus on expanding the data set, incorporating more variables, and integrating contextual data to further improve sleep quality analysis and support real-time health interventions.

The substantial increase in the human population dramatically strains food supplies. Farmers need healthy soil and natural minerals for traditional farming, and production takes a little longer time. The soil-free farming method known as vertical farming (VF) requires a small land and consumes a very small amount of water than conventional soil-dependent farming techniques. With modern technologies like hydroponics, aeroponics, and aquaponics, the notion of the VF appears to have a promising future in urban areas where farming land is very expensive and scarce. VF faces difficulty in the simultaneous monitoring of multiple indicators, nutrition advice, and plant diagnosis systems. However, these issues can be resolved by implementing current technical advancements like artificial intelligence (AI)-based control techniques such as machine learning (ML), deep learning (DL), the internet of things (IoT), image processing as well as computer vision. This article presents a thorough analysis of ML and IoT applications in VF system. The areas on which the attention is concentrated include disease detection, crop yield prediction, nutrition, and irrigation control management. In order to predict crop yield and crop diseases, the computer vision technique is investigated in view of the classification of distinct collections of crop images. This article also illustrates ML and IoT-based VF systems that can raise product quality and production over the long term. Assessment and evaluation of the knowledge-based VF system have also been outlined in the article with the potential outcomes, advantages, and limitations of ML and IoT in the VF system.