Studies on trust oriented machine Learning based mitigating schemes For ssdf attack in cognitive radio Networks

Abstract

Spectrum scarcity limits the fulfillment of demand for the proliferation in wireless services. Cognitive Radio Network (CRN) deploys its intelligence in solving the issue of spectrum scarcity. The CRN enables the unlicensed users (secondary users -SU) to utilize the licensed user s spectrum without causing interference when their spectrum is idle. The CRN autonomously detects the white spaces in the spectrum and allocates the free spectrum to the SU s by its functionalities such as Spectrum Sensing, Spectrum analysis, Spectrum mobility and sharing. The ability to reconfigure its parameters according to the user s necessity by lively monitoring the environment proves the intelligence of the CRN. Several SU s employ Cooperative Spectrum Sensing (CSS), by sharing their spectrum sensing results to cooperatively detect the presence of the Primary User (PU) Signal. Also CSS takes the advantage of overcoming the deterioration of detection of PU signal by fading and shadowing. newlineSecuring the CRN becomes a prime factor due to the wireless and reconfigurability nature of the CRN. Also the complete functionality of CRN depends on the spectrum sensing results, consequently any threat affecting the sensing reports would obliterate the system. Spectrum Sensing Data Falsification attack (SSDF) is the threat that affects the spectrum sensing results of the CRN, with the objective to maliciously grab the spectrum or to selfishly occupy the spectrum there by exempting genuine SU s from using the spectrum. Mitigating SSDF attack requires a cognitive approach as different types of attackers can cause this attack. newline