Robust tracking of mobile primary user based on interval analysis in cognitive radio networks

To improve the spectrum utilization efficiency, secondary users (SUs) in cognitive wireless networks should be able to not only detect the spectral holes in time and frequency domains, but also exploit the spatial spectrum opportunities for dynamic spectrum access. In this paper, accurate and robust tracking of mobile-primary user is formulated as a Bayesian filtering problem with the shadow-fading estimation. Then, a novel interval analysis based Rao-Blackwellized particle filtering algorithm is proposed to obtain the approximately optimal estimates of the mobile primary user's location and the shadow-fading gain with only a small number of particles. Since the interval analysis is used to model the target states, the proposed method can obtain a satisfied tracking performance with low computational complexity. Simulation results validate the effectiveness of the proposed method.