This study aims to identify the most suitable IoT communication protocol for a digital twin architecture, focusing on meeting the demands of real-time data transmission with stability and security. The research methodology involves a comparative performance analysis of MQTT over TCP, MQTT over WebSocket, and HTTP protocols, evaluating their latency, connectivity stability, and IoT application feasibility using an ESP32 microcontroller and DHT22 sensor setup. The experimental results demonstrate that MQTT over TCP achieves the lowest average latency at 290.5 ms, while MQTT over WebSocket exhibits more stable latency profiles with a 193.2 ms standard deviation; HTTP, despite its broader compatibility, showed higher average latency of 342.6 ms with a 307.9 ms standard deviation. These findings provide valuable insights for digital twin implementations, enabling developers to make informed protocol selections based on specific requirements, whether prioritizing real-time performance through MQTT protocols or emphasizing system compatibility and simplicity through HTTP.

Industrial Internet of Things (IIoT) deployments increasingly rely on low-cost microcontrollers and single-board computers to stream operational telemetry for monitoring, control, and predictive maintenance, yet the canonical “TLS-to-broker” model does not protect message content from a compromised or curious MQTT broker. This study therefore designs and implements a practical, application-layer end-to-end (E2E) encryption pipeline spanning an ESP32 data client (C++/mbedTLS), an untrusted MQTT broker, and a Raspberry Pi gateway (Python/PyCryptodome) using AES-256-GCM with Additional Authenticated Data (AAD). Sensor measurements are serialized as compact JSON, encrypted and authenticated on the ESP32, framed into a binary record, Base64-encoded for MQTT payload carriage, and verified/decrypted only at the gateway. Experiments on ESP32-WROOM-32 and Raspberry Pi 4 show an average ESP32 packet-preparation latency of 41.754 ms (JSON 1.0 ms; AES-GCM 29.5 ms; Base64 11.2 ms), robust rejection of ciphertext tampering and unauthorized devices via MAC verification and whitelist checks, and 99.72% decrypt-and-store success over a one-hour run (718/720 messages). These results indicate that commodity IIoT hardware can support practical and replicable E2E confidentiality and integrity without sacrificing operational throughput, while eliminating the MQTT broker as a de facto man-in-the-middle.

This paper analyzes the dynamics of cyberattacks and fraud techniques using machine learning algorithms for Industrial Internet of Things (IIoT) systems. Special attention is paid to information security issues of digital twins within the framework of the Industry 4.0 concept. Various machine learning techniques, such as logistic regression, random forest, and the nearest neighbor method, are considered for classifying attacks in IIoT systems, taking into account the problem of data imbalance inherent in rare attacks. The use of data balancing and cross-validation methods can improve the accuracy of the models and minimize the number of false positives. The results demonstrate that the random forest model outperforms other methods in terms of accuracy, confirming its effectiveness for IIoT security and digital twin protection. Additionally, feature selection and optimization techniques were explored to enhance model performance further. The study also highlights the importance of real-time threat detection, which is crucial for maintaining the integrity and resilience of IIoT environments. Future research is planned to investigate more sophisticated methods, including deep learning, to improve attack detection and classification in the context of Industry 4.0. The integration of federated learning and adaptive AI-driven security measures may also offer promising solutions to emerging cyber threats.

The development of a wearable sensor device integrated into the internet of things (IoT) infrastructure is presented, with functionality aimed at continuous measurement of the user's physiological parameters and their intelligent processing for real-time stress level assessment. The system enables continuous monitoring of physiological parameters, allowing early detection of stress signals and supporting adaptive behavioral responses. The hardware platform is designed to consolidate various biomedical sensors, enabling continuous acquisition and intelligent processing of physiological data in real time. During testing, heart rate (HR) ranged from 68 to 89 beats per minute (bpm), respiratory rate varied from 11 to 15 breaths per minute, and skin conductivity ranged from 63 to 77 µS. Acquired physiological data were uploaded to a cloud-based infrastructure to enable advanced processing and analysis. The system achieved an overall stress detection accuracy of 87%, and signal stability remained high even under changing conditions. The proposed wearable solution demonstrates strong potential for use in healthcare, education, and occupational environments. It also offers scalability through the integration of intelligent algorithms and additional sensor modules.

The integration of Advanced Metering Infrastructure (AMI) provides high-resolution electrical data, essential for enhancing industrial efficiency and monitoring equipment health. However, the utility of this data is frequently compromised by anomalies, underscoring the necessity for robust, automated detection methodologies. This study benchmarks three distinct categories of machine learning models: a statistical baseline (SARIMA), an unsupervised classifier (Isolation Forest), and a deep learning reconstruction model (LSTM-Autoencoder). The evaluation was conducted using a multivariate dataset acquired from bakery manufactory equipment, employing a synthetic anomaly injection framework with a 5% contamination rate. The results indicate significant challenges in accurately detecting anomalies within this dataset. The SARIMA model achieved the highest average F1-Score (0.256), slightly outperforming the Isolation Forest (0.233), while the LSTM-Autoencoder performed the poorest (0.110). Critically, all models exhibited extremely low precision (ranging from 0.074 to 0.204), indicating an unacceptably high rate of false positives. The findings suggest that standard configurations of these models struggle to differentiate between true anomalies and the inherent variability of industrial operations, highlighting the need for advanced optimization and feature engineering for practical deployment.

The proliferation of Internet of Things (IoT) devices in financial markets has created distributed ecosystems where cryptocurrency exchanges, trading platforms, and market data providers operate as autonomous edge nodes generating massive volumes of sensitive financial data. Collaborative machine learning across these distributed financial IoT nodes faces fundamental challenges: institutions possess valuable proprietary data but cannot share it directly due to competitive concerns, regulatory constraints, and trust management requirements in decentralized networks. This study presents a privacy-preserving federated learning framework tailored for distributed financial IoT systems, combining differential privacy with Shamir secret sharing to enable secure collaborative intelligence across blockchain-based cryptocurrency trading networks. We implement per-layer gradient clipping and Rényi differential privacy composition to minimize utility loss while maintaining formal privacy guarantees in edge computing scenarios. Using 5.6 million orderbook observations from 11 cryptocurrency pairs collected across distributed exchange nodes, we evaluate three data partitioning strategies simulating realistic heterogeneity patterns in financial IoT deployments. Our experiments reveal that federated edge learning imposes 9–15 percentage point accuracy degradation compared to centralized cloud processing, driven primarily by data distribution heterogeneity across autonomous nodes. Critically, adding differential privacy (ε = 3.0) and cryptographic secret sharing increases this degradation by less than 0.3 percentage points when mechanisms are calibrated appropriately for edge devices. The framework achieves 62–66.5% direction accuracy on cryptocurrency price movements, with confidence-based execution generating 71–137 basis points average profit per trade. These results demonstrate the practical viability of privacy-preserving collaborative intelligence for distributed financial IoT while identifying that the federated optimization gap dominates privacy mechanism costs. Our findings offer architectural insights for designing trustworthy distributed systems in blockchain-enabled financial IoT ecosystems.