Comparative analysis of efficiency of algorithms for detection of signals in cognitive resource allocation in mobile network

The analysis of the efficiency of parametric algorithm optimal signal reception, which was built on the Wald criteria; nonparametric algorithm, built on the mathematical formalism of neural networks; nonparametric algorithm, built on the mathematical apparatus of fuzzy logic; nonparametric algorithm, built on the signs, the Wilcoxon rank test. Analysis of the effectiveness of the proposed algorithms performed in Matlab environment using simulation. As a criterion of the effectiveness of selected probability of error. Obtain plots of the probability of error of the ratio of signal power to noise power ratio (SNR). The analysis showed that at low SNR values from -10 dB to -7dB best indicators has an algorithm built on the mathematical apparatus of the Kohonen neural network. With the increase in SNR (above -7 dB) the smallest probability of error is noted when using the algorithm Wald. But, although the Wald algorithm has better efficiency, it requires a priori information about signal parameters. When the SNR is above 0 dB lower probability of error among nonparametric algorithms has an algorithm built on the mathematical apparatus of fuzzy logic. This is slightly higher probability of errors observed in the algorithm, built on the Kohonen network. The most likely error has detection algorithm based on the characters, the Wilcoxon rank test.