Spectrum Sensing Based on Time-frequency Dispersion Analysis
Wenxiu Zheng · Journal of Information and Computational Science · 2014
A novel spectrum sensing algorithm is proposed based on time-frequency dispersion analysis. Timefrequency ridge line is flrstly calculated of a received data’s mutual correlation to form a series of timefrequency dots. Time-frequency dispersion function is subsequently deflned to describe the dispersion level of the time-frequency dots according to the fact that signal energy gather around the carrier frequency and noise energy disperse evenly among all frequencies. And the dispersion threshold is set to make an existence decision of primary user’s signal. Simulations indicates: the time-frequency dispersion is robust to noise power uncertainty, receiver’s bandwidth and behaves stability with enough data length; the proposed algorithm is efiective both for normal and fading channel and maintains good detecting performance at lower SNRs.