Global outliers detection in wireless sensor networks: A novel approach integrating time‐series analysis, entropy, and random forest‐based classification.

Abstract

Wireless sensor networks (WSNs) have recently attracted greater attention worldwide due to their practicality in monitoring, communicating, and reporting specific physical phenomena. The data collected by WSNs is often inaccurate as a result of unavoidable environmental factors, which may include noise, signal weakness, or intrusion attacks depending on the specific situation. Sending high-noise data has negative effects not just on data accuracy and network reliability, but also regarding the decision-making processes in the base station.Anomaly detection, or outlier detection, is the process of detecting noisy data amidst the contexts thus described. The literature contains relatively few noise detection techniques in the context of WSNs, particularly for outlier-detection algorithms applying time series analysis, which considers the effective neighbors to ensure a global-collaborative detection. Hence, the research presented in this article is intended to design and implement a global outlier-detection approach, which allows us to find and select appropriate neighbors to ensure an adaptive collaborative detection based on time-series analysis and entropy techniques. The proposed approach applies a random forest algorithm for identifying the best results. To measure the effectiveness and efficiency of the proposed approach, a comprehensive and real scenario provided by the Intel Berkeley Research Laboratory has been simulated. Noisy data have been injected into the collected data randomly. The results obtained from the experiment then conducted experimentation demonstrate that our approach can detect anomalies with up to 99% accuracy.

A Literature Review on Outlier Detection in Wireless Sensor Networks

“…However, the specific definition can vary depending on the context and the methodologies upon which ODTs are based [3,16]. Among the most used definitions in the literature is [27] "An observation, which deviates so much from other observations as to arouse suspicions that it was generated by a different mechanism" [4,20,25,28,29]; and a little more recently that of [16] "Those measurements that significantly deviate from the normal pattern of sensed data" [21,25,[29][30][31].…”

“…In Ref. [28], a technique is presented that combines time series analysis, entropy, and classification using random forests. Finally, the technique in [52] uses Generative Adversarial Networks (GANs), an unsupervised learning approach, to detect outliers in WSNs, implementing two neural networks and autoencoders trained through the Adam optimizer.…”

“…The technique proposed in [28] integrates multiple approaches and techniques, which could increase its complexity and computational requirements. In addition, the proposed technique does not yet address the problem of detecting malicious or faulty nodes, which can negatively affect the accuracy of outlier detection.…”

MGAD: Mutual Information and Graph Embedding Based Anomaly Detection in Multivariate Time Series

“…Graph neural networks (GNNs) are widely used in research to identify anomalies in multivariate time data [41]. In order to discover anomalous periods through prediction and reconstruction, gated recurrent units are utilized to record patterns in the time series, while graph attention networks are utilized to learn correlations across multivariate time series [42].…”

A Microscopic Platoon Stability Model Using Vehicle-to-Vehicle Communication

“…Thus, the adoption of C-V2X will significantly improve the connectivity of the platoon due to the advanced 5G technology. In this paper, we used the traditional IEEE 802.11p [36,37], which may lead to the issues mentioned above. The proposed model is a base for a stable platoon and can be extended to C-V2X in the future, aiming to overcome the problems related to the dis-connectivity of VANETs.…”