The goal of algorithms for joint beamforming and power control in communication systems with antenna arrays is to minimize the transmission power while keeping the signal-to-interference-plus-noise ratio (SINR) above a minimum acceptable level. To meet this goal, one appealing approach is to first use a training sequence to adaptively obtain the beamforming coefficients and then, in a second step, to update the transmission power. However, such a power update often changes the beamforming scenario abruptly, leading to a nonstationary behavior that may greatly impair the performance of the adaptive beamforming algorithm. To overcome this problem, a novel approach for joint beamforming and power control is proposed here. Such an approach is based on a continuous iterative update of the transmission power that allows controlling the undesirable nonstationary behavior of the beamforming scenario, thereby significantly improving the performance of the joint beamforming and power control. Simulation results are shown aiming to confirm the effectiveness of the proposed approach. A new joint beamforming and power control strategy is proposed.The proposed strategy uses training sequences of mobile communication systems.In contrast to other strategies, the new one is based on a continuous power update.The new strategy allows for a very effective joint beamforming and power control.Simulation results show the effectiveness of the proposed power control strategy.