A Network Intrusion Detection Approach at the Edge of Fog

Azarkasb, Seyed Omid; Kashi, Saeed Sedighian; Khasteh, Seyed Hossein

doi:10.1109/csicc52343.2021.9420618

Cited by 1 publication

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The benefits of NN include avoiding explicit detection rule development, identifying known malformed assaults, and the capacity to control noise and outlier data. 67,72 The utilization of NN eliminates the explicit development of rules, identifies known malformed assaults, and controls noise and outlier data, besides reducing the overhead generated through persistent resource employment in fog. 72,91 ML techniques have several drawbacks.…”

Section: Resultsmentioning

confidence: 99%

“…67,72 The utilization of NN eliminates the explicit development of rules, identifies known malformed assaults, and controls noise and outlier data, besides reducing the overhead generated through persistent resource employment in fog. 72,91 ML techniques have several drawbacks. To begin with, their performance is strongly reliant on the resilience of the feature engineering approach used, limiting their stability.…”

Section: Resultsmentioning

confidence: 99%

“…Azarkasb et al 72 began by describing the network topology for locating detector nodes. The diagnostic node analyzer was created using the self-organizing map.…”

Section: Neural Network-based Idssmentioning

confidence: 99%

See 2 more Smart Citations

A deep and systematic review of the intrusion detection systems in the fog environment

Yi¹,

Yin²,

Darbandi

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Fog computing has arisen to complement cloud computing, offering a cost-effective architecture to power the Internet of things. Fog computing is a network computing and service paradigm that expands cloud computing and services to the network's edge to solve the cloud's fundamental shortcomings, like a lack of location awareness, mobility, and latency. Besides, fog devices may have security issues because fog nodes are near end-users and have limited computational power. As the linked devices in the Internet of things grow exponentially, so does the potential for cyber-attack. Intrusion detection systems have been developed as a feasible tool for detecting and preventing harmful activity in this vein. This work used a systematic literature review to examine intrusion detection system approaches in a fog environment. By categorizing intrusion detection systems in the literature, we produce a review. An intrusion detection system's characteristic is early identification of an attack and the decrease of false alarms as a result. Some of the issues that these systems' designers will face are also discussed. Recall rate, detection rates, detection delay, and detection precision were the parameters that received the most attention. This report will serve as a roadmap for future network intrusion detection studies in fog computing. 1• Numerous fog nodes are managed and controlled by a cloud computing center.

show abstract

Section: Resultsmentioning

confidence: 99%