Exploiting Multichannel Diversity in Spectrum Sharing Systems Using Optimal Stopping Rule

Date Added: Apr 2012
Format: PDF

This paper studies the problem of exploiting multichannel diversity in a spectrum sharing system, where the Secondary User (SU) sequentially explores channel state information on the licensed channels with time consumption. To maximize the expected achievable throughput for the SU, the authors formulate this problem as an optimal stopping problem, whose objective is to choose the right channel to stop exploration based on the observed signal-to-noise ratio sequence. Moreover, they propose a myopic but optimal rule, called one-stage look-ahead rule, to solve the stopping problem.