Enhancing Cybersecurity in Wireless Sensor Networks: Innovative Framework for Optimized Data Aggregation

The various cyberattacks in wireless sensor networks (WSNs) have made confidentiality and data integrity as crucial principles in data aggregation. Therefore, several applications are presented to control the sharing of data and information as well as the associated cybersecurity aspects that must be preserved during data transfer. Most cybersecurity breaches that occur these days are categorized as cyberattacks. The WSN’s resource-constrained architecture makes cybersecurity lapses and insider attacks possible. This study proposes a novel technique named multi-objective pigeon-inspired optimal long short-term memory (MPI-OLSTM) networks to develop the data aggregation in cybersecurity model. Initially, the WSN-detection systems (WSN-DS) dataset is collected and pre-processed using min-max normalization. For extracting features, the principal component analysis (PCA) is employed. The model’s predictive power is assessed using the following metrics: accuracy (96.5%), precision (92.3%), and recall (90.4%). The findings demonstrate that, in comparison to existing techniques, our approach yielded more accurate results.