Spectrum sensing algorithm based on extremum and geometric average of the power spectral

The detection performance of spectrum sensing algorithms should be improved under low signal to noise radio, unknown carrier offset and noise uncertainty. The maximum value of signal’s power spectral is estimated by the part sample-averaging of power spectral to reduce the effect of frequency offset of signal on spectrum sensing performance. In order to remove the influence of noise and preserve primary user’s signal features, the ratio of the difference of the average value of maximum and minimum power spectral to the power spectral geometric average is used as the decision statistics. The detection threshold is deducted, from which we can find that the proposed algorithm is insensitive to noise uncertainty. The simulation results in additive white Gaussian noise channel and Rayleigh fading channel show that the proposed algorithm is better than the existed algorithms based on power spectral, which reduces the influence of unknown carrier offset and noise uncertainty to the performance of spectrum sensing algorithms. The algorithm can detect practical signal very effectively.