Intelligent Secure Data Aggregation in WSNs

Abstract

The paper discusses the problem of secure data aggregation in wireless sensor networks (WSNs) - a procedure that is of critical importance for reducing energy consumption, minimizing transmission overhead, and thus prolonging network lifetime. Due to the limited computational and energy resources of WSN nodes, traditional aggregation methods often fail to perform effectively in dynamic heterogeneous environments. With such a context taken into consideration, this study emphasizes the potential of artificial intelligence techniques, such as neural networks, genetic algorithms, and fuzzy logic, to enable adaptive aggregation approaches tailored to environmental and network-specific parameters. Furthermore, the integration of fuzzy logic, genetic algorithms, and artificial neural networks into a hybrid system leverages the strengths of each approach, resulting in enhanced adaptability and accuracy of the aggregation process. As part of the investigation, a fuzzy inference system (FIS) model was developed that incorporates attributes such as energy, current load, distance to the base station, and trust level. The model was implemented in Matlab using the Fuzzy Logic Designer toolbox. To further improve system performance, a genetic algorithm was applied to optimize membership functions. In the final phase, the model was transformed into an adaptive neurofuzzy inference system (ANFIS) which was trained using simulated data within Matlab. The simulation results demonstrate that the proposed hybrid approach ensures flexible, robust and energy-efficient control of the data aggregation process under dynamically changing conditions in which WSNs operate.