Novel protocol for enhancing the speed of dynamic spectrum sensing in cognitive radio

Cognitive radio systems are capable of sensing the prevailing environmental conditions and automatically adapting its operating parameters in order to enhance system and network performance. Spectrum sensing is an important enabling technology for future opportunistic spectrum sharing scenarios. In this paper, a novel protocol is proposed to determine the maximum number of samples required to obtain the desired Probability of Detection (Pd) for a given SNR. In energy detection, the Probability of Detection has varied against the number of samples (N) for different values of SNR. It is observed that, for a particular SNR, the Probability of Detection increases continuously with the increasing number of samples. However after a certain value of the number of samples, the increase in the Probability of Detection becomes insignificant. In order to optimize the number of samples required to obtain the desired Probability of Detection (Pd), first, the adaptive threshold is to be fixed with CFAR theory then a set of curves are plotted between (Pd) Vs N with varying SNR values. From the plots the local maxima points thereafter the Pd improvement is insignificant are noted. These points are referred to as (Pd-sat) points with respect to various SNR values and are representing the maximum number of samples required to obtain the desired Probability of Detection (Pd), for a given SNR. This simulation work is carried out using MATLAB. The proposed protocol is verified using sin, OFDM and DVB-T signals as inputs. The results are tabulated and plotted. From the results it is observed that the proposed procedure is working satisfactorily. Niemen Pearson detector is used to calculate the probability of detection. This protocol seems to be simple and does not require solving complex mathematical models representing the dynamic spectrum sensing.