PEAK POWER MINIMIZATION IN SYMBOL-LEVEL PRECODING FOR COGNITIVE MISO DOWNLINK CHANNELS

This paper proposes a new symbol-level precoding scheme at the cognitive transmitter that jointly utilizes the data and channel information to reduce the effect of nonlinear amplifiers, by reducing the maximum antenna power under quality of service constraint at the cognitive receivers. In practice, each transmit antenna has a separate amplifier with individual characteristics. In the proposed approach, the precoding design is optimized in order to control the instantaneous power transmitted by the antennas, and more specifically to limit the power peaks, while guaranteeing some specific target signal-to-noise ratios at the receivers and respecting the interference temperature constraint imposed by the primary system. Numerical results show the effectiveness of the proposed scheme, which outperforms the existing state of the art techniques in terms of reduction of the power peaks.