Real-Time Detection and Isolation of Malicious Aggregator Nodes in Wireless Sensor Networks Using Trust-Based Evaluation: A Systematic Literature Review

The Wireless Sensor Networks (WSNs) have become a key enabler of critical applications in industries like industrial automation, smart infrastructure monitoring, etc. One of the basic problems of the WSN operation is that an aggregator node is able to add false data to the system that can be spread to the base station and distort the system-level decision making. In this paper, a systematic literature review (SLR) is conducted, which aims at summarizing mechanisms for real-time detection and isolation of malicious aggregator nodes from 47 peer-reviewed publications published from IEEE Xplore, ACM Digital Library, ScienceDirect and SpringerLink in the last 10 years (2018-2026). After presenting the data integrity, real-time and aggregator-node attack inclusion and exclusion criteria aligned with PRISMA, five main technical categories are discussed: homomorphic encryption, trust-based management, iterative statistical filtering, witness-based verification and AI/ML-based intrusion detection. A cross-study synthesis demonstrates that there is always a trade-off among the three key measures of detection latency, energy overhead, and Byzantine resilience; none of the currently proposed approaches can satisfy all three constraints. Four concrete research gaps are identified and a future research direction, the Lightweight Adaptive Trust-Integrated Aggregation (LATIA) framework is proposed.