Spectrum Sensing with LAD-Based Methods

Opportunistic spectrum usage would enable enhancing the efficiency of existing and emerging wireless communication systems. One of the key issues related to those systems is spectrum opportunity estimation. In this paper, we are using a technique utilized earlier in narrowband signal detection, namely the localization algorithm based on double-thresholding (LAD), for sensing the existence of primary user signals in a cognitive radio systems. The LAD method requires no a priori information on the primary user statistics and it has a low computational complexity that will enable a low-cost real-time implementation. The LAD method is able to estimate the number of narrowband signals and their characteristics, including bandwidth and power. A simplified version of the LAD method which uses normalized thresholds (NT) as well as an enhancement of the scheme that uses adjacent cluster combining (ACC) are proposed. Simulation results show that the simplified version of the LAD method is useful in the considered situations, and the enhanced version of the LAD method improves the performance of the LAD and LAD NT methods significantly.