Spectrum Sensing based on Energy Detection using MATLAB_Simulink

Spectrum sensing is the major tasks of cognitive radio for utilizing spectrum resources efficiently thereby identifying and making use of spectrum holes is the objective of sensing. Energy detection technique is adopted to compare the performance of different modulated signals in poor and good SNR. Probability of detection and false alarm are the metrics used to identify the number of primary users over a bandwidth of 5MHz. False alarm is kept constant to increase the detection probability. Simulink model for detections based on FFT and Welch periodogram are implemented. Modulated signals having high data rates can occupy more number of users thereby increasing the usage of spectrum. Also increase in noise should not affect the detection of primary users and reduce the probability of misdetection. Receiver operating characteristic curve is used to analyze the detections with respect to false alarm probability for different ranges of SNR. Noise uncertainty and SNR wall which can limit the performances are also taken into consideration. A unique approach to energy detection based on FFT shows increase in the sensing in the form of spectrum estimation with four primary users and the noise is assumed to be additive white Gaussian. Neyman Pearson hypothesis is considered for determining the presence and absence of users. Noise is the factor which decreases the performance of energy detector in case of blind sensing which can be limited with the help of highly modulated signals. The growing demand for spectrum makes cognitive radio an important technology to be focused in which the sensing process can reduce the interference caused to the primary users.