Monotone Split and Conquer for Anomaly Detection in IoT Sensory Data

Dang, Thien-Binh; Le, Duc-Tai; Nguyen, Tien-Dung; Kim, Moonseong; Choo, Hyunseung

doi:10.1109/jiot.2021.3073705

Cited by 17 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Time-series data [43,44,45,46,47,48,49,50] Smart city [51,52,53,54] Monitoring machinery health [55,5] Robotics and manufacturing [56,57,58] Detection applications for IoT sensors [59,60] Other industrial or manufacturing [61,62,63,64] Surveillance and video [65,66,67] General-purpose frameworks [68,69,70] Network and communication frameworks [71,72,73,74] User security and privacy frameworks [75,76,77] Other frameworks [78,79] Network traffic in IoT [80,81,82] Device and infrastructure security [83,21,84,85,86,87] Data transport security [88,…”

Section: Application Category Referencementioning

confidence: 99%

“…The method trains supervised models (Random Forests, Naive Bayes, and Neural Networks) to detect anomalies and cites a 30x reduction in computational time. Dang et al [64] proposes a novel monotone split and Conquer (MSC) strategy to detect short and long forms of anomalies. The MSC model has an offline training phase and an online detection phase.…”

Section: Other Industrial or Manufacturing Applicationsmentioning

confidence: 99%

See 1 more Smart Citation

IoT anomaly detection methods and applications: A survey

2022

View full text Add to dashboard Cite

Section: Application Category Referencementioning

confidence: 99%

Section: Other Industrial or Manufacturing Applicationsmentioning

confidence: 99%

IoT anomaly detection methods and applications: A survey

2022

View full text Add to dashboard Cite

“…In terms of IoT, there are a huge number of sensors generating large volumes of multidimensional data, where data anomalies caused by system errors or malicious attacks do harm to the application effectiveness or even make the system break down [16]- [18]. As a crucial challenge in IoT, anomaly detection has been widely investigated in system security and monitoring management fields, including Intrusion Detection Systems (IDSs) [16], [17], [19]- [21], fraud monitoring [22], [23], data-based decision [24]- [26] and so on. Conceptually, anomaly detection refers to digging out the abnormal data that is inconsistent with the general data in the viewpoint of given rules or features, which is also regarded as a binary-classification issue for normal data and abnormal data.…”

Section: A Backgroundmentioning

confidence: 99%

From MIM-Based GAN to Anomaly Detection:Event Probability Influence on Generative Adversarial Networks

Sheng¹,

Fan²

2022

Preprint

View full text Add to dashboard Cite

In order to introduce deep learning technologies into anomaly detection, Generative Adversarial Networks (GANs) are considered as important roles in the algorithm design and realistic applications.In terms of GANs, event probability reflected in the objective function, has an impact on the event generation which plays a crucial part in GAN-based anomaly detection. The information metric, e.g. Kullback-Leibler divergence in the original GAN, makes the objective function have different sensitivity on different event probability, which provides an opportunity to refine GAN-based anomaly detection by influencing data generation. In this paper, we introduce the exponential information metric into the GAN, referred to as MIM-based GAN, whose superior characteristics on data generation are discussed in theory. Furthermore, we propose an anomaly detection method with MIM-based GAN, as well as explain its principle for the unsupervised learning case from the viewpoint of probability event generation. Since this method is promising to detect anomalies in Internet of Things (IoT), such as environmental, medical and biochemical outliers, we make use of several datasets from the online ODDS repository to evaluate its performance and compare it with other methods.

show abstract

“…Therefore, a technology capable of continuously collecting data in the observation area without defects has been studied as a significant issue in recent years. For example, in areas where severe natural disasters have occurred or areas where people cannot access, such as conflict areas, essential data can be collected using various Internet of Things (IoT) devices [2][3][4]. Small sensors can be mounted and distributed in unmanned aerial vehicles such as drones, but it is not easy to evenly place small IoT devices throughout the observation area through spraying from drones.…”

Section: Introductionmentioning

confidence: 99%

Novel Sensing Hole Recovery with Expanded Relay Node Capability

Kim¹,

Lee²

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

The occurrence of 'sensing holes' not only hinders seamless data collection but also leads to misinterpretation of information in certain areas under extensive data analysis. In order to overcome this, various sensor relocation strategies have been proposed, but the existing relocation strategies revealed problems such as the ping-pong, shaded area, network disconnection, etc. This paper conducted research on relocation protocols in a distributed environment that is very suitable for real-world situations and efficiently recovering the problem of sensing holes. First, a simulation was performed on the distribution of the shaded area for data collection, which is a problem with the existing representative relocation protocol. After that, a data collection capability was newly added to the relay node, which had been in charge of only communication between cluster zones so far, and with this additional functionality, the performance of the revised sensor relocation algorithm was dramatically improved to overcome the existing problems. In addition, the performance and validity of the proposed algorithm were verified through various simulations.

show abstract

Monotone Split and Conquer for Anomaly Detection in IoT Sensory Data

Cited by 17 publications

References 30 publications

IoT anomaly detection methods and applications: A survey

IoT anomaly detection methods and applications: A survey

From MIM-Based GAN to Anomaly Detection:Event Probability Influence on Generative Adversarial Networks

Novel Sensing Hole Recovery with Expanded Relay Node Capability

Contact Info

Product

Resources

About