Evolutionary computation technique enhancing the performance of cognitive radio networks with energy harvesting

The intervention of evolutionary computation methods in studying energy harvesting (EH) techniques in multiple-input-single-output cognitive radio networks (MISO-CRNs) is carefully investigated. In this work, a secondary user (SU) with multiple antennas harvests energy from a hybrid base station (HBS) in addition to harvesting from the primary user (PU) transmission during the downlink time-slot. Hence, the SU transmits its data on the uplink over Nakagami-m fading channels. For the underlay paradigm, a specific tolerable PU interference limits the SU total transmission power. The HBS does not have previous information about the SU battery level. A closed-form expression for the SU exact outage probability is derived and simplified to its asymptotic formula for the high signal-to-noise-ratio (SNR) region. Closed-form expressions for the system average symbol error probability (ASEP) and ergodic capacity are obtained. The particle swarm optimization (PSO) algorithm is used to find the optimal SU transmission power to minimize the sophisticated formula of the exact outage probability. Simulations and numerical results reveal the effect of the critical system parameters on the overall system performance and validate the derived mathematical analysis. Moreover, the optimum solution of the power optimization problem shows noteworthy enhancement of the system performance compared to the traditional equally distributed model. (C) 2020 Elsevier B.V. All rights reserved.