Adaptive and Channel-Aware Detection of Selective Forwarding Attacks in AdHoc Networks

Abstract

- We propose a Channel-aware name System with adaptive detection threshold (CRS-A) to detect selective
forwarding attacks in WSNs. The CRS-A evaluates the info forwarding behaviors of device nodes, in keeping with the
deviation of the monitored packet loss and therefore the calculable traditional loss. To optimize the detection accuracy of
CRS-A, we tend to on paper derive the best threshold for forwarding analysis, that is accommodative to the time varied
channel condition and therefore the calculable attack possibilities of compromised nodes. Moreover, associate attacktolerant knowledge forwarding theme is developed to collaborate with CRS-A for exciting the forwarding cooperation of
compromised nodes and up the info delivery quantitative relation of the network. Intensive simulation results
demonstrate that CRS-A will accurately notice selective forwarding attacks and determine the compromised device
nodes, whereas the attack-tolerant knowledge forwarding theme will considerably improve the info delivery quantitative
relation of the network. We will extend our investigation into wireless circumstantial network with mobile device nodes,
wherever the detection of selective forwarding attacks becomes tougher, since the conventional packet loss rate is
additional fluctuant and troublesome to estimate attributable to the quality of device nodes.