Currently, cognitive radio systems (CRSs) are of great interest to improve the efficiency of using the available spectrum. The spectrum sharing technique to share the spectrum between a primary transmitter (PT) and a secondary transmitter (ST) is influential in CRSs. However, when reusing the PT spectrum, the ST transmits energy and decoded data to the secondary users (SUs), and providing power to the energy receivers (ERs) also interferes with the primary users (PUs). One method used to increase the power delivered to the ERs, limit the interference to the PUs and increase the total channel capacity to the SUs is to use the weighted minimum mean squared error (WMMSE) method. This method combines increasing the channel capacity to the SUs and minimizing the error for the transmission channels. It will not focus on solving error reduction but rather on increasing the channel capacity through the precoder design at the STs. The proposed algorithm combines using the eigenvectors of the known channel matrix at the ST to generate beams; the distributed power value for each beam is calculated based on the Lagrangian operator combined with the Karush–Kuhn–Tucker (KKT) conditions to maximize capacity. This new method allows increased capacity compared to systems using WMMSE by other methods such as weighted sum rate (WSR), harmonic mean rate (HMR), or proportional fairness (PF) in different situations such as increasing the number of SUs, ERs, or PUs or limiting interference to given PUs..

Cognitive Radio Systems primary transmitter primary users secondary transmitter seconday users energy receivers transmit precoders