Soft Computing - A step towards building Secure Cognitive WLAN

  S.C.Lingareddy1                               Dr B Stephen Charles2                   Dr Vinaya Babu3               Kashyap Dhruve4 

Asst. professor and HOD, CSE, Principal Professor of CSE, Technical Director
KNS Institute of Technology, Stanley Stephen College of Engg, Director of Admission Dept. Planet-i Technologies Bangalore, India. Kurnool, India. JNTU, Hyderabad, India. Bangalore, India.

Abstract— Wireless Networks rendering varied services has not only become the order of the day but the demand of a large pool of customers as well. Thus, security of wireless networks has become a very essential design criterion. This paper describes our research work focused towards creating secure cognitive wireless local area networks using soft computing approaches. The present dense Wireless Local Area Networks (WLAN) pose a huge threat to network integrity and are vulnerable to attacks.
In this paper we propose a secure Cognitive Framework Architecture (CFA). The Cognitive Security Manager (CSM) is the heart of CFA. The CSM incorporates access control using Physical Architecture Description Layer (PADL) and analyzes the operational matrices of the terminals using multi layer neural networks, acting accordingly to identify authorized access and unauthorized usage patterns.

Keywords- Cognitive Networks , Back Propogation , Soft Computing, IEEE 802.11, WLAN Security,Cognitive Framework Architecture, Multilayered Feedforward Neural Network (MFNN), Physical Architecture Description Layer(PADL), Cognitive Security Manager(CSM). I. INTRODUCTION Wireless Networks have become an integral part of the Information Technology (IT) Infrastructure especially Wireless Local Area Network (WLAN). The Institute of Electrical and Electronics Engineers (IEEE) have described certain standards and specification for efficient communication over the wireless medium IEEE 802.11[26]. WLAN’s have seen a tremendous growth because of their ability to provide flexible and mobility options to the user. WLAN’s are now an integral part of both Enterprise Network and Public Networks. With such dense WLAN’s available to user’s, security is indeed a major concern. The security features provided in the current deployed WLAN’s are vulnerable [1]. Cognitive Networks also known as smart networks [2] could be a solution to this security and data integrity concern. Cognitive Networks are recognized for their self aware, self management and self healing properties. Wireless networks have been studied as complex, heterogeneous and dynamic environments and cognition of these are still under research [3]. Cognitive networks could be used to improve resource management [4],[17], quality of service (QoS)[5], security and access control[6]. We propose to introduce a CFA to secure WLAN’s. Computational Intelligence is defined as the study and design of intelligent components. To impart computational intelligence we propose to use soft computing techniques. Soft Computing encapsulates various several intelligence imparting technologies including fuzzy logic, neural networks, probabilistic computing, artificial immune systems etc. We intend to use multilayer neural networks in our CFA.
In this paper we put forward the CFA where we would use multiple layers of neural networks to impart intelligence to the proposed framework. The intelligence parameters of the CFA are analyzed in the Experimental Setup. II. RELATED WORK WLANS are one of the most lucrative and fast growing deployments for connectivity options for networks. The WLANS are established by Access Points (AP). The AP that are provided by the manufactures or service providers. Based on the literature provided to the users they feel that the AP’s are secure which is not the fact [7]. The AP’s provide security in multiple ways like WEP [8], [22], WPA [9] ,TKIP[10], etc which are vulnerable and could be easily rendered ineffective. Even the authentication schemes provided in the AP, based on the user node hardware are ineffective. The current AP’s provide Ethernet MAC based filtering for access control but this too was found to be insecure [11]. It is very clear that the WLAN has been known for its non secure nature [7]. Many protocols have been developed with a hope to provide better security [27]. However, much research is ongoing to solve this security issue [10][13][14].
Based on our research we believe that these WLANS could be made intelligent and the access control mechanism could be improved to negate these security deficiencies. Much work has been done towards securing wireless networks based on cognitive approaches. When we speak of cognition terms like software d

…(Full text truncated)…