The Impact of Integrating Artificial Intelligence Techniques Into Data Fusion for Wireless Sensor Networks: A Systematic Literature Review

Abstract

Wireless Sensor Networks (WSNs) are pivotal in modern applications such as environmental monitoring, smart cities, industrial automation, and healthcare systems. However, the large volume of data generated by sensor nodes often leads to redundancy, noise, and energy inefficiency. To address these challenges, data fusion techniques have been increasingly enhanced by Artificial Intelligence (AI) to improve data quality, reduce communication overhead, and enable intelligent decisionmaking. This paper presents a systematic literature review (SLR) of AI-based data fusion methods in WSNs published between 2015 and 2025. Using well-defined search strategies and inclusion/exclusion criteria across major digital libraries (IEEE Xplore, Scopus, web of science and Science Direct), 33 relevant studies were selected and analyzed. The review categorizes existing approaches based on the type of AI techniques used (e.g., machine learning, deep learning, hybrid methods), the fusion levels (sensor, node, gateway), and their application domains. Furthermore, we identify common challenges such as energy constraints, scalability, real-time processing, and model interpretability. This review highlights current research gaps and proposes directions for future work, particularly in the integration of lightweight AI models and edge computing for efficient and scalable data fusion in WSN environments.