Evaluation of Micro-slots Spectrum Sensing under Noise uncertainty and PU Random Access

To achieve the spectrum utilization objective of cognitive radio, the unauthorized user (SU) should avoid interference to the authorized user (PU) when the SU accesses the licensed channel and should access the licensed channel when it is actually available. In a normal scenario, traditional spectrum sensing determines the channel status when all samples of the sensing slot contain the PU signal. However, PU can utilize the licensed channel indulgently where the PU has the right to access or leave the channel anytime during the sensing interval. However, the indulgent access behavior of PU affects the detection performance of traditional spectrum sensing techniques since they do not take the behavior into the account. Later, modern techniques address the indulgent access behavior issue by exploiting micro-slots detection concepts. However, they did not evaluate the detection performance under noise uncertainty environment where the noise uncertainty is the major concern of the spectrum sensing. Therefore, in this paper, we determine the generalization character of the three modern spectrum sensing, micro-slotted dual-stage spectrum sensing (MDMS), split-half spectrum sensing (SPHS) and elastic fusion dual-stage spectrum sensing (EFDS), where the generalization can be met when the spectrum sensing presents reliable detection under various condition of spectrum usage of PU with noise uncertainty environment. As a result, EFDS can achieve the generalization character since it is capable of adjusting the required number of PU detected adjacent micro-slots.