MMSE Based Block Diagonalization for Cognitive Radio MIMO Broadcast Channels

In this paper, we present a linear precoder design for cognitive radio (CR) multi-user multiple-input multiple-output (MU-MIMO) downlink systems where unlicensed secondary users (SUs) can simultaneously utilize the same spectrum used by a licensed primary user (PU). Although a zero-forcing block diagonalization (ZF-BD) precoder is extended to the CR network with the PU, a transmit power boost problem occurs. Therefore, we propose a regularized BD precoder method based on the minimum mean-squared error (MMSE) criteria subject to the interference power constraint under a predetermined threshold for the PU. As a result, the proposed CR-MMSE-BD scheme improves the signal-to-interference-plus-noise ratio at each SU's receiver, compared to the ZF-BD based method. The simulation results demonstrate that the proposed algorithm outperforms the ZF based technique for CR MU-MIMO downlink systems.