Particularly in Internet of Things (IoT) scenarios, the rapid growth and diversity of network traffic pose a growing challenge to network intrusion detection systems (NIDs). In this work, we perform a comparative analysis of lightweight machine learning models, such as logistic regression (LR) and k-nearest neighbors (KNNs), alongside other machine learning models, such as decision trees (DTs), support vector machines (SVMs), multilayer perceptron (MLP), and random forests (RFs) with deep learning architectures, specifically a convolutional neural network (CNN) coupled with bidirectional long short-term memory (BiLSTM), for intrusion detection. We assess these models’ scalability, performance, and robustness using the NSL-KDD and UNSW-NB15 benchmark datasets. We evaluate important metrics, such as accuracy, precision, recall, F1-score, and false alarm rate, to offer insights into the effectiveness of each model in securing network systems within IoT deployments. Notably, the study emphasizes the utilization of lightweight machine learning models, highlighting their efficiency in achieving high detection accuracy while maintaining lower computational costs. Furthermore, standard deviation metrics have been incorporated into the accuracy evaluations, enhancing the reliability and comprehensiveness of our results. Using the CNN-BiLSTM model, we achieved noteworthy accuracies of 99.89% and 98.95% on the NSL-KDD and UNSW-NB15 datasets, respectively. However, the CNN-BiLSTM model outperforms lightweight traditional machine learning methods by a margin ranging from 1.5% to 3.5%. This study contributes to the ongoing efforts to enhance network security in IoT scenarios by exploring a trade-off between traditional machine learning and deep learning techniques.

Fires and fire hazards in industrial environments pose a significant risk to safety, infrastructure, and the operational community. The need for real-time monitoring systems capable of detecting fires early and transmitting alerts promptly is crucial. This paper presents a fire monitoring system utilizing lightweight communication protocols, a multi-hop wireless network, and anomaly detection techniques. The system leverages Message Queue Telemetry Transport (MQTT) for efficient message exchange, the ESP-NOW for low-latency and reliable multi-hop wireless communications, and a radio environment map for optimal node placement, eliminating packet loss and ensuring robust data transmission. The proposed system addresses the limitations of traditional fire monitoring systems, providing flexibility, scalability, and robustness in detecting fire. Data collected by ESP32-CAM sensors, which are equipped with pre-trained YOLOv5-based fire detection modules, are processed and transmitted to a central monitoring server. Experimental results demonstrate a 100% success rate in fire detection transmissions, a significant reduction in latency to 150ms, and zero packet loss under REM-guided configuration. These findings validate the system’s suitability for real-time monitoring in high-risk industrial settings. Future work will focus on enhancing the anomaly detection model for greater accuracy, expanding scalability through additional communication protocols, like LoRaWAN, and incorporating adaptive algorithms for real-time network optimization.