A Worst-Case Robust Distributed Power Allocation in OFDM-based Cognitive Radio Networks

This paper considers the problem of decentralized power allocation for throughput optimization under the power constraint and interference limits in OFDM-based cognitive radio(CR) network, where second users(SUs) are allowed to share the sub-carriers of the OFDM-based primary network. Worst-case approach is used to discuss the imperfect channel state information between users. A robust distributed power allocation is proposed, where the power allocation among the (SUs) is formulated as a robust non-cooperative game. Then, the game is rewritten as a variational inequality problem to obtain conditions guaranteeing the uniqueness of the NE of the game. A robust iterative water-filling algorithm is proposed to obtain the of cognitive user power allocation solution. The simulation results show that robust distributed power allocation is more suitable for the actual channel model and avoid interference to primary users(PUs).