Outlier Detection in Sensor Data Using Machine Learning Techniques for IoT Framework and Wireless Sensor Networks: A Brief Study

Ghosh, Nimisha; Maity, Krishanu; Paul, Rourab; Maity, Satyabrata

doi:10.1109/icaml48257.2019.00043

Cited by 25 publications

(17 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, some works deal with preparing data to be used more effectively by ML algorithms. The results show that the better the quality of the data will be the results obtained in the classification [17][18][19].…”

Section: Introductionmentioning

confidence: 93%

Intelligent IoT security monitoring based on fuzzy optimum-path forest classifier

Souza

Medeiros

et al. 2022

Soft Comput

View full text Add to dashboard Cite

The detection of intrusions in IoT networks is essential to maintain the availability and integrity of data transmitted and generated by devices connected to these networks. This is primarily when the data originates from critical activities, such as activities in the military, financial, industrial, and health sectors. Machine learning techniques have been adopted to create ways to detect or improve the accuracy of existing models for automatic intrusion detection. However, it is difficult to find in the literature an accurate intrusion detection technique in an IoT environment, as there are different types of attacks that can happen in different ways. Therefore, to solve this problem, this work proposes applying Fuzzy OPF (Optimum-Path Forest) as a new detection algorithm for any threat that escapes the regular traffic of an IoT network. We evaluate our proposed approach by using five different ML algorithms: Linear Discriminant Analysis, Support Vector Machine, Bayes, K-Nearest Neighbors, and Optimum-Path Forest. Experimental results analysis showed that our proposed model outperforms well-known algorithms in the literature regarding the Accuracy, Recall, and F1 metrics.

show abstract

Section: Introductionmentioning

confidence: 93%

Intelligent IoT security monitoring based on fuzzy optimum-path forest classifier

Souza

Medeiros

et al. 2022

Soft Comput

View full text Add to dashboard Cite

show abstract

“…There are several works with different approaches to detecting anomalies. Consequently, Works such as Gosh et al [28], Gaddam et al [29], Choi et al [1]. and Cook et al [30] presented earlier state-of-the-art studies showing the most relevant techniques used for detecting anomalies.…”

Section: Related Workmentioning

confidence: 99%

Hybrid Anomaly Detection Model on Trusted IoT Devices

Rosero-Montalvo

István

Tözün

et al. 2023

IEEE Internet Things J.

View full text Add to dashboard Cite

Most machine learning proposals in the Internet of Things (IoT) are designed and evaluated on pre-processed datasets, where data acquisition and cleaning steps are often considered a black box. In addition, IoT environments have numerous challenges related to acquiring data from sensors, where sensitive data can be threatened by malicious users who seek to interfere with the communication channel or storage. Additionally, sensor data can also be affected by noise. Therefore, differentiating the type of threat/anomaly requires additional energy and computational resources. We propose to carry out data cleaning/anomaly techniques on the IoT device itself, not in the cloud servers but closer to the data source. Therefore, the IoT device sends trusted data to the Cloud. Among the benefits of this is the considerable reduction in the cost of implementation due to less movement of data between IoT devices and the Cloud. Consequently, we define three anomaly detection steps using smoothing filters, unsupervised learning, and deep learning techniques (i.e., hybrid model) to detect different variations of anomalies and threats while focusing on a small computational/memory footprint. The deployment of the hybrid model on AVR, Tensilica, and ARM microcontrollers showed that the last ones are an adequate target to implement the model because they best satisfy the necessary hardware requirements. The proposed model consumes 50KB of Flash and 12KB of RAM and processes data locally, achieving a bandwidth reduction of 60%. Finally, the hybrid model was tested in external datasets.

show abstract

“…A detailed taxonomy of machine learning approaches for IoT is provided in [11] where a description on how machine learning algorithms can be applied to IoT smart data and what are the characteristics of IoT data in the real world are discussed. Several papers propose to use machine learning for anomaly detection or security issues (e.g., [12,13]) while others use machine learning for specific application purposes such as healthcare [14,15], traffic analysis [16,17], resource management [18] and modelling routing strategies in opportunistic networks [19]. The different application contexts and the potentialities of the use of machine learning are described in [14] where a review on the use of these techniques in different IoT application domains for both data processing and management tasks is provided.…”

Section: Related Workmentioning

confidence: 99%

“…From network data it is possible to extract users' behaviours as shown in the paper by Tao et al [16] in which users' activity features are extracted with regard to inter-arrival times and packets' traffic size and their relationships are investigated. The use of classification methods has been suggested in [17] to check thoroughly the quality of data collected by IoT or wireless sensor networks. The authors show that classification is the most extensively adopted learning method for detecting outliers in IoT and wireless sensor networks.…”

Section: Related Workmentioning

confidence: 99%

Predicting LoRaWAN Behavior: How Machine Learning Can Help

et al. 2020

View full text Add to dashboard Cite

Large scale deployments of Internet of Things (IoT) networks are becoming reality. From a technology perspective, a lot of information related to device parameters, channel states, network and application data are stored in databases and can be used for an extensive analysis to improve the functionality of IoT systems in terms of network performance and user services. LoRaWAN (Long Range Wide Area Network) is one of the emerging IoT technologies, with a simple protocol based on LoRa modulation. In this work, we discuss how machine learning approaches can be used to improve network performance (and if and how they can help). To this aim, we describe a methodology to process LoRaWAN packets and apply a machine learning pipeline to: (i) perform device profiling, and (ii) predict the inter-arrival of IoT packets. This latter analysis is very related to the channel and network usage and can be leveraged in the future for system performance enhancements. Our analysis mainly focuses on the use of k-means, Long Short-Term Memory Neural Networks and Decision Trees. We test these approaches on a real large-scale LoRaWAN network where the overall captured traffic is stored in a proprietary database. Our study shows how profiling techniques enable a machine learning prediction algorithm even when training is not possible because of high error rates perceived by some devices. In this challenging case, the prediction of the inter-arrival time of packets has an error of about 3.5% for 77% of real sequence cases.

show abstract

Outlier Detection in Sensor Data Using Machine Learning Techniques for IoT Framework and Wireless Sensor Networks: A Brief Study

Cited by 25 publications

References 22 publications

Intelligent IoT security monitoring based on fuzzy optimum-path forest classifier

Intelligent IoT security monitoring based on fuzzy optimum-path forest classifier

Hybrid Anomaly Detection Model on Trusted IoT Devices

Predicting LoRaWAN Behavior: How Machine Learning Can Help

Contact Info

Product

Resources

About