Most common types of malicious user

Most common types of malicious user attacks are: 1) PUE attack, 2) Spectrum Sensing Data False (SSDF) [5] attack, 3) using vacant spectrum opportunity or transmitting noise in links (Jamming and Game theory), 4) secondary emulation and participating in spectrum sharing. In this scheme, we mainly focus on the security problem from PUE attacks in CRN. In this attack MUs emulate the PUs' transmitting signal and mislead the legitimate SUs to give up the spectrum band. For instance, if a MU is to be recognized as a PU, a TV broadcaster with specific modulation, rate and power, should transmit the same signal with exactly PU’s features. This scenario would not be endured for a long time because MU wants to transmit its own data and features. In recent approaches, MUs have been detected by knowing positions of other SUs and PUs, while the mobility of users makes the detection unsuccessful [6]. MU can be recognized by Feature Selection Algorithms (FSAs) like k-means and graph theory in the absence of training data.