A spectrum sensing algorithm based on Kolmogorov-Smirnov two-sample test

In this article, a spectrum sensing method based on Kolmogorov-Smirnov two-sample test is proposed. The spectrum sensing problem is formulated as a two-sample goodness of fit test problem. A Kolmogorov-Smirnov statistic is constructed based on the received signal samples and previously collected noise samples, and the decision on the signal presence is made by comparing the value of the statistic with a preset threshold. A theoretical analysis on the lower bound of detection probability for the proposed method is given. The method does not require any information about signal, channel and noise distribution. Simulation results are presented to verify the performance of the method.