A deep learning approach for effective intrusion detection in wireless networks using CNN
- PDF / 662,276 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 72 Downloads / 332 Views
(0123456789().,-volV)(0123456789(). ,- volV)
METHODOLOGIES AND APPLICATION
A deep learning approach for effective intrusion detection in wireless networks using CNN B. Riyaz1 • Sannasi Ganapathy1
Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Security is playing a major role in this Internet world due to the rapid growth of Internet users. The various intrusion detection systems were developed by many researchers in the past to identify and detect the intruders using data mining techniques. However, the existing systems are not able to achieve sufficient detection accuracy when using the data mining. For this purpose, we propose a new intrusion detection system to provide security in data communication by identifying and detecting the intruders effectively in wireless networks. Here, we propose a new feature selection algorithm called conditional random field and linear correlation coefficient-based feature selection algorithm to select the most contributed features and classify them using the existing convolutional neural network. The experiments have been conducted for evaluating the proposed intrusion detection system that achieves 98.88% as overall detection accuracy. The tenfold crossvalidation has been done for evaluating the performance of the proposed model. Keywords Convolutional neural network Conditional random field Correlation coefficient Feature selection Classification and intrusion detection system
1 Introduction The rapid growth of computer networking technology is facilitating the comfort in the businesses, organizations and social communities. Simultaneously, different types of Internet security threats kept developing due to the regular rise of various vulnerabilities and attacking techniques. Hence, some security systems should be used for preventing the attacks and to provide the confidentiality as well as the availability of resources and integrity for the Internet communications. To determine and restrict the mischievous network traffic, intrusion detection system (IDS) has become the most important network security solution (Liao et al. 2013).
Communicated by V. Loia. & B. Riyaz [email protected] Sannasi Ganapathy [email protected] 1
School of Computing Science and Engineering, Vellore Institute of Technology, Chennai, Tamilnadu, India
The IDS is supervising the real network flow and gathers the network packets in the infrastructure. The IDS is used for inspecting the network packets for detecting malicious activities and blocks the mischievous network from intruders or attacks (Aburomman and Reaz 2017). IDS is capable of classifying the records into five types such as normal, Probe, Denial of Service (DoS), User to Root (U2R) and Root to Local (R2L) according to their behaviour in network (Ganapathy et al. 2013). Generally, IDS is categorized into two types, namely misuse-based IDS and anomaly-based IDS (Ganapathy et al. 2013). The misuse-based IDSs are identified the attacks by comparing the intrusion activity with the predefined signat
Data Loading...
