Precoder design in cognitive radio networks with channel covariance information

In cognitive radio communications, the coexistence of a secondary network is allowed as long as the interference power from the secondary network measured at the primary user (PU) is kept below a given threshold. In this paper, we consider a communication scenario for cognitive radio networks where the channel state information (CSI) between the secondary base station (SBS) and the PU is perfectly known at the SBS transmitter whereas only the covariance information is known for the channel of the SBS-secondary user (SU) link at the SBS transmitter. A closed-form precoder design method is proposed for the purpose of maximizing the upper bound of the expected achievable rate of the SU. Even though the proposed scheme uses only partial CSI of the SBS-SU link, the simulation results show that the expected achievable rate of the proposed scheme approaches to that of a conventional scheme with full CSI as the channel correlation increases.

► Precoder design problem using a partial channel state information (CSI) is treated in cognitive radio networks.
► We assume the CSI between a secondary base station (SBS) and a primary user is perfectly known to the SBS.
► We also assume only the channel covariance information of the SBS-secondary user (SU) link is known to the SBS.
► We propose a closed-form solution for a precoder design that maximizes the upper bound of the expected achievable rate of the SU.