摘要

In view of the deficiencies of the traditional spectrum detection algorithms, a novel signal detection algorithm-spectrum detection based on two-dimensional features in cognitive radio networks is proposed. It puts forward a new instantaneous characteristics parameter, which is called the average value of the absolute value of the zero-center normalized instantaneous energy. Then, a two-dimensional vector is constituted by the box dimension and the new instantaneous characteristics parameter. Finally, by taking this two-dimensional vector as the test statistics for spectrum detection, the novel algorithm not only shows a strong detection performance, but also can accurately identify the type of signal with a high recognition ratio. The numerical results show that, in 3dB environment, the algorithm proposed can completely correct detect the signal existed and identify the modulation type of this signal.

全文