Rde-Moga: Automatic Selection Of Rate-Distortion-Energy Control Points For Video Encoders Using Muti-Objetive Genetic Algorithm

Controlling energy consumption of video encoders is acomplex multi-objective optimization problem of great im-portance. In this work we propose the RDE-MOGA, an multi-objective genetic algorithm capable of finding energeticallyefficient configurations for the HEVC encoder and replacingthe current sensitivity analysis methodologies in the develop-ment of energy controllers. The utilization of our algorithmimproved its efficiency in 60% whereas increasing the rangeof achievable reductions of the controller in at least 50%. Fur-thermore, the algorithm proved capable of sustaining 30% en-ergy reduction at a cost of 3.45 BD-BR loss.