The Performance Analysis And Verification Of Spectrum Sensing Method Base On Energy Detection

Nowadays, the mobile wireless communications services are widely used around the world, the demand for spectrum resources faces enormous increase. For this situation, people often use spectrum sensing, so it can solve the problem of the shortage of the spectrum resources effectively. Energy detection is the most widely used for detecting the spectrum sensing technologies. Because it can be performed without knowing a priori information of the authorized user, and the lower complexity and easy to implement. So the energy detection technique is widely used, so it is important for studying the technology. In energy detector for spectrum sensing, the decision threshold is a predefined value for deciding.Firstly, it introduces the concept of spectrum sensing technology, and it compares the advantages and disadvantages of the various detection methods in spectrum sensing technology. Especially for the noise factor in energy detection algorithms, it makes a brief presentation instructions. In the practical energy detectors by discovered, noise power has uncertainty due to thermal noise and interference, which should be considered for energy detection. And it can affects the detection performance. In addition, it makes the energy detection techniques applied to GNU Radio and USRP experimental platform of software radio in this paper.Then, it based on Newman-Pearson criterion energy detection technique and it deduces the complete expression of the threshold in energy detection and uncertainty noise on the basis of the existing model of the noise power uncertainty. The relations between the decision threshold and the relevant factors(sampling points-. the probability of detection and false alarm probability)are derived and verified by simulation in MATLAB. At the same time, it discusses the issues about the detection threshold during the SNR wall caused by noise uncertainty.Finally, it based on GNU Radio and USRP experimental platform of cognitive radio, the detection in the time domain is transformed to the frequency domain by the FFT. It tests the energy detection techniques within the frequency range of790M to810M. It compares the threshold value in energy detection technology under different environmental impacts of signal power. So it can be determined the relationship between the threshold and the SNR. It is found that the results on GNU Radio and USRP experimental platform and the theoretical simulation by MATLAB are consistent. It tests the spectrum sensing method successfully and provides a verification platform for the application of cognitive radio technology.