Abstract:Significant growth of smart home devices is adopted, which provides security, convenience, and energy efficiency for users, in recent years. As an instance, consider a secured smart camera that detects movements of unauthorized objects, whereas the fire accidents can be detected by smoke sensors. Though, a surface for new cyber threats is some of the most recent examples that are open up in this field. Furthermore, recent examples also show that the distributed denial of service (DDoS) attacks are performed by misusing the smart devices which are hacked violating the privacy rules. In this proposed work, the application of machine learning in an environment of smart homes is explored so that the anomalous activities that occur can be identified. The sensor data at the network level is trained using a model based on the Markov chain known as hidden Markov model (HMM) and is created using smart devices and multiple sensors from a testbed. The model generated using HMM achieves 97% accuracy while detecting potential anomalies where attacks are indicated. In this approach, we construct the model and differentiate the analysed results with existing techniques. The starting step for securing IoT networks is intrusion detection and the next step is the intruders prediction which provides an active defence against the incoming attacks. The model employs an algorithm that does not depend on specific domain knowledge. The work has achieved an improvement in prediction accuracy of 5% for an alert category over the current variable length methods of Markov chain intrusion prediction, as they provided information more for a possible defence. The DDoS attack is considered as a coordinated attack mainly carried out based on a large scale depending on whether the target system’s resources or services are available. Thus, the novel approach also describes a novel machine learning-based technique using an algorithm known as variational dynamic Bayesian algorithm which helps to obtain a HMM with the number of parameters and model states which are optimized for the prediction of a DDoS attack. This procedure conquers the speed of a moderate combination HMM approach.

Modeling program behaviors by hidden Markov models for intrusion detection

DPHMM: Customizable Data Release with Differential Privacy Via Hidden Markov Model.

Profiling program behavior for anomaly intrusion detection based on the transition and frequency property of computer audit data

Hidden Semi-Markov Model for Anomaly Detection

Robust Algorithm for Attack Detection Based on Time‐varying Hidden Markov Model Subject to Outliers

Worm Intrusion Alarm Modeling Based on Network Traffic Character

Multi-agent based hybrid Intrusion detection system

Measuring and Enhancing the Transferability of Hidden Markov Models for Dynamic Travel Behavioral Analysis

Hidden Markov Model-Based Attack Detection for Networked Control Systems Subject to Random Packet Dropouts

IT Intrusion Detection Using Statistical Learning and Testbed Measurements

Multimode Process Monitoring Approach Based on Moving Window Hidden Markov Model

Profiling Program and User Behaviors for Anomaly Intrusion Detection Based on Non-Negative Matrix Factorization

Intrusion Alert Prediction Using a Hidden Markov Model

Method for Masquerade Intrusion Detection Based on HMM and Genetic Algorithm

A model for intrusion detection system using hidden Markov and variational Bayesian model for IoT based wireless sensor network

A variable-length model for masquerade detection

Novel Method for Anomaly Detection of User Behavior Based on Shell Commands and DTMC Models

Masquerade Detection Based on Shell Commands and High-Order Markov Chain Models

Markov Graph Model Computation and Its Application to Intrusion Detection

A Markov Model of System Calls Sequence and Its Application in Anomaly Detection