Use AI to predict required communication volume and allocate resources efficiently OKI and Tohoku University|電経新聞

Use AI to predict required communication volume and allocate resources efficiently OKI and Tohoku University

仮想PONシステムと実験の様子(Virtual PON system and experiment)

As part of the “Research and Development Project for Strengthening the Infrastructure of Post-5G Information and Communication Systems” commissioned by NEDO, OKI and Tohoku University are working to efficiently operate passive optical network (PON) systems used in optical internet services. They have developed a “virtualization resource control technology” that uses AI to predict the required amount of communication and allocate resources efficiently.
This is the world’s first successful experiment in predicting total traffic volume using AI and autonomously increasing and decreasing wavelength resources using an OSS-based virtual PON. If we translate the effect of this technology into power consumption, we can expect a reduction of more than 20% compared to existing methods.