Assessing IoT Intrusion Detection Computational Costs When Using a Convolutional Neural Network

Abstract

IoT systems face vulnerabilities due to their data processing requirements and resource constraints. With 13 billion connected devices globally, this research investigates the economic viability of AI-based intrusion detection systems (IDSs), specifically analyzing the automation costs of implementing a Convolutional Neural Network (CNN) with Long Short-Term Memory (LSTM) for classifying malicious sensor traffic. This study introduces an innovative framework that evaluates six distinct architectural components of CNN and LSTM: image input processing, convolutional layer operations, max pooling layer functionality, fully connected layer characteristics, softmax output activation, and class determination mechanisms. The framework employs six metrics: matrix size, feature vector number, input vector size, output vector size, and number of runs for dual data points. Experiments on the IoT-23 dataset showed our proposed CNN model outperformed LSTM, achieving 93% accuracy for binary classification and 96% for multi-class classification. The trained CNN demonstrated predictable resource utilization with increasing classification complexity, providing a framework for quantifying IoT IDS costs. The proposed framework provides a systematic methodology for evaluating machine learning classifiers in IoT environments, using quantitative metrics to assess implementation and operational costs, enabling data-driven selection of optimal security solutions based on specific deployment constraints.