International Journal of Engineering Research

Asfa Sadaf Albadri ¹, Chandrakant Naikodi ¹, L. Suresh ²

Affiliations
1 Dept. of CSE, Cambridge Institute of Technology, Bangalore, IN
2 Cambridge Institute of Technology, Bangalore, IN

Abstract

Variations from the norm in detected information streams demonstrate the spread of malicious attacks, hardware failure and software corruption among the diverse nodes in a remote sensor system. These variables of node infection can influence generated and approaching information streams bringing about high odds of off base information, deluding packet interpretation, wrong choice making and serious correspondence interruption. This issue is unfavorable to constant applications having stringent quality of service (QoS) prerequisites. The detected information from other uninfected regions may likewise get stuck in a infected area ought to no earlier option game plans are made. Albeit a few existing techniques (BOUNDHOLE and GAR) can be utilized to relieve these issues, their execution is limited by a few confinements, predominantly the high danger of falling into steering circles and association in pointless transmissions. This paper gives an answer for by-pass the infected nodes progressively utilizing a twin moving balls procedure furthermore redirect the packets that are caught inside the recognized region. The recognizable proof of infected nodes is finished by adjusting a Fuzzy data clustering approach which categorizes the nodes in light of the portion of peculiar information that is recognized in individual information streams. This data is then utilized as a part of the proposed by-passed Routing(BPR) which pivots two balls in two headings at the same time: clockwise and counter-clockwise. The primary node that hits any ball in any heading and is uninfected, is chosen as the next hop. We are likewise worried with the approaching packets or the packets on-the-fly that might be influenced when this issue happens. Other than taking care of both of the issues in the current techniques, the proposed BPR strategy has significantly enhanced the considered QoS parameters as appeared by very nearly 40 percent expansion in the general execution. We can also effectively detect the malicious nodes that attempt to launch gray hole attacks and malicious nodes are detected using a reverse tracing technique. Any detected malicious node is kept in a black list so that all other nodes that participate to the routing of the message are alarmed to quit speaking with any node in that list and we can implement cooperative load balancing.

Keywords

BOUNDHOLE, GAR, BPR, Fuzzy Data Clustering, Gray Hole Attacks, Reverse Tracing Technique, Wireless Sensor Networks.

Full Text