Research On Spectrum Sensing Algorithm Based On Adaptive Threshold

Due to the growing claim of wireless communication services,spectrum has been a limited resource.So that prompted many regulatory bodies from different parts of world to consider cognitive radio as a finest solution for spectrum scarcity and inefficiently usage of that insufficient spectrum.Among the vital functions of CR,spectrum sensing is regarded to be the foremost important function key of it.In numerous literatures,spectrum sensing using energy detection scheme was preferred by many researchers due to its easiness and requiring no prior knowledge about PU’s transmission as well as taking short time during sensing phase.However,energy detector does not give out finest results due to variation of noise power.Also,performing sensing activity employing individual CR may not give valid output because of unseen terminal problems,shadowing and multipath fading.To mitigate these problems,in this thesis we have proposed adaptive threshold aimed at minimising error decision probability in deciding on the existence of PU signal as well as mitigating the problem of poor performance of energy detector in lowest SNR conditions.Furthermore,we have proposed a system model where multiple CRs have been considered in sensing activity.MRC(Maximum Ratio Combining)receiver diversity scheme have been considered to fuse all sensed information at the FC and energy detection method to quantify the amount of energy the combined signal holds.Our simulation results have clearly shown that performing spectrum sensing employing adaptive threshold out competes fixed conventional one in both cooperative and in non-cooperative.Also proposed model is capable of recognizing the occurrence of PUs even in stumpy SNR regions of-10dB.Finally,we have presented graphically that adaptive threshold achieves high spectrum efficiency when multiple CRs have been employed than when single CR has been used.