Enhancing IoT Data Security with a Hybrid Cryptographic Framework
R. Yuvarani, R Mahaveerakannan · 2024
The Internet of Things (IoT) is experiencing rapid growth due to technological advancements, but its security remains a significant challenge. This research proposes a dual approach to enhance IoT security. First, a deep learning-based malware detection system is developed using Long Short-Term Memory (LSTM) neural networks. This system identifies malicious nodes and predicts attack types by analyzing trust values, contextual information, and network data. Second, an Improved Elliptic Curve Cryptography (IECC) algorithm is introduced to secure data transmission between IoT devices. The proposed system demonstrates high accuracy (95%), low error rate (5%), and superior performance compared to existing encryption methods. By combining advanced malware detection with robust encryption, this research offers a comprehensive solution to protect IoT devices and data.