Deep neural network based anomaly detection in Internet of Things network traffic tracking for the applications of future smart cities

Abstract

An anomaly exposure system's foremost objective is to categorize the behavior of the system into normal and untruthful actions. To estimate the possible incidents, the administrators of smart cities have to apply anomaly detection engines to avert data from being jeopardized by errors or attacks. This article aims to propose a novel deep learning‐based framework with a dense random neural network approach for distinguishing and classifying anomaly from normal behaviors based on the type of attack in the Internet of Things. Machine learning algorithms have the improbability to explore the performance, compared with deep learning models. Distinctively, the examination of deep learning neural network architectures achieved enhanced computation performance and deliver desired results for categorical attacks. This article focuses on the complete study of experimentation performance and evaluations on deep learning neural network architecture for the recognition of seven categorical attacks found in the Distributed Smart Space Orchestration System traffic traces data set. The empirical results of the simulation model report that deep neural network architecture performs well through noticeable improvement in most of the categorical attack.

Characterizing Network Anomaly Traffic with Euclidean Distance-Based Multiscale Fuzzy Entropy

“…Stochastic processes like Hidden Markov Model and Conditional Random Field were also frequently applied in detection of traffic anomaly [12,13]. Due to the success of deep-learning technologies in image processing and natural language processing, they have been intensively studied in network intrusion detection [14,15], network traffic tracking [16], and network traffic abnormal behavior detection [17]. Besides, time-series density analysis [18], wavelet [19], principal components analysis [20], and ensemble learning technologies [21] have been extensively investigated in network anomaly detection.…”

“…k,m (i, j)/r)); (16) end for (17) Calculate the average similarity between vector Y (τ) k,m (i) and the other vectors…”

“…Characterization of network anomaly traffic is one of the key technologies commonly used to model and detect network anomalies and then to raise the cybersecurity awareness capability of network administrators. e existing approaches of network anomaly detection can be mainly classified into six categories [1]: classification-based methods [2][3][4], clustering-based methods [5][6][7][8][9], statistical methods [10,11], stochastic methods [12,13], deep-learning-based methods [14][15][16][17], and others [18][19][20][21].…”

Performance Estimation in V2X Networks Using Deep Learning-Based M-Estimator Loss Functions in the Presence of Outliers

“…Maimó et al [27] studied DL's performance for abnormality discovery in 5G networks. Reddy et al [28] examined anomaly detection by DNN in tracking IoT traffic for future smart-city applications.…”

Special issue on deep network based industrial Internet of Things applications

“…Therefore, to avert data from being jeopardized by errors or attacks, it is requisite to have anomaly detection engines for smart cities' IoT networks. In the proposed work by Reddy et al, 7 the foremost objective is to study the behavior of the system to identify attacks against IoT networks and by categorizing them into normal and abnormal actions. Deep Learning meets the requisite complex and nonlinear relations without any premodel, and this leads to exploring the performance when compared with traditional machine learning models.…”