Primary emitter localization using smartly initialized Metropolis-Hastings algorithm

Abstract

The knowledge of the primary emitter location is important in cognitive radio networks as it is required to determine the exclusion region of the primary network. We show that interpolation based localization techniques do not provide accurate primary emitter localization; however they can provide significant complexity reduction when their estimates are used to initialize more accurate iterative localization techniques. In this paper, we generated interference maps using low complexity interpolation techniques and provided their coarse estimates to initialize a Metropolis-Hastings (MH)based localization algorithm. Our simulation results show that smart initialization of the MH algorithm eliminates tedious parameter tuning process and achieves significantly better localization performance than randomly initialized MH algorithm at a fraction of iterations.