Spectrum sensing based on linear self-signal elimination for opportunistic spectrum access

We discuss the spectrum sensing problem which is crucial in opportunistic spectrum access. An In-band spectrum sensing method based on linear self-signal elimination is proposed. In this method, quiet period is not needed so that transmission performance can be improved without the interruption of on-going transmission. According to the AR model of the wireless fading channel, the adjacent symbols are linear weighted to eliminate the self-signal. Further the optimal detector under Neyman-Pearson criterion is used on the residual signal after self-signal elimination to provide better detection performance. Finally, performance is evaluated with simulations which show that our method can provide better performance in the IEEE802.22 environment.