Abstract
Digital communication is provided with an effective communication platform to share and transfer information. The emergence of the Cyber-Physical System (CPS) is a platform incorporated with electronic devices that enables the services through a digital platform. The considerable challenges of this system are security issues, abnormality, and service failure. Hence, the requirement of providing an effective system, which should be overcome these issues. This paper analyzes these problems and providing the paradigm in terms of enhanced communication paradigm, specifically propose Energy Aware Smart Home (EASH) framework. With this work, the problem in communication failures and types of network attacks are analyzed in EASH. With the utilization of the machine learning technique, the abnormality sources of the communication paradigm are differentiated. To evaluate the performance, we analyze the proposed work based on its accuracy, performance, and efficiency. Hence, we obtain better results especially the result shows an 85% accuracy rate. In the future, we try to enhance a high accuracy rate for further development.
Introduction
The advancement in modern technology enables effective communication to every field, in specific the Cyber-Physical System is a novel platform to provide a better platform to share and transfer information from one end to another via the different communicational channel. This technology enables advancement in communication transferring platforms; therefore, the development of the economy is reached a high destination. However, security and resiliency is still challenging and should be considered for security enhancement processes. The two considerable factors [1] of communication failure are security and component failures. The hackers or intruders enable malicious activity towards the CPS system; meanwhile, the development of it is tremendous and omnipresent in modern societies. Hence, security failure is a serious topic due to the abnormal behaviors of the system; however, the implications of these are not the same as other systems. Therefore, the development of new technology is to tackle these issues in terms of minimizing the attacks, providing security service, and protecting existing services by controlling abnormal behaviors. In this process, the parameter specification is the complex task of differentiation which needs the research on the CPS system based on its specific components [2-3], before incorporating with a holistic strategy.
Conclusion
This article provides findings on the differentiation of labor between the anomalies affecting each system. Anomalies that are discussed in this paper may attack data or network attacks. The relationship between the type of anomaly and its impact on the system communication channel differentiation operator was studied and well-appointed. They are used for the approach based ML for classification of differentiation. The outcomes indicated the utilization of algorithms in the supervised machine learning was a positive methodology by separate between the harmed class and forceful with a serious extent of precision. The incorrect classification of cases with the same impact on the network and then analyzed, either in experimental settings (real-time simulation and testing), good for two classes of anomalies and characteristics examined. Our analysis shows that the ranking results can be further improved by adding or removing attributes from the dataset descriptive.
