Development of a data collection and storage system for remote monitoring and detection of security threats in the enterprise

Abstract

As the Industrial Internet of Things (IIoT) expands, maintaining a high level of security and reliability is becoming increasingly important for uninterrupted operations. Encryption (TLS/SSL and AES-256) and intrusion detection and prevention systems (IDPS) are essential. In addition, the platform uses neural network algorithms, namely long short-term memory (LSTM) and hybrid CNN-LSTM models, to identify anomalies in real time, which contributes to a rapid response to potential failures or cyber threats. Through the use of model compression and explainable AI (XAI) techniques, the architecture adapts to a variety of industrial scenarios without compromising performance or transparency, helping industry professionals strengthen security measures and improve real-time anomaly detection in the ever-evolving IIoT landscape.