Sustainability and Power Outage-aware Placement of Edge Computing in NextG RANs
Kuheli Sai, David Tipper · 2024
The advent of Next Generation (NextG) Radio Access Networks (RANs) necessitates the strategic geographical placement of edge computing infrastructure to meet the demands for low latency, high bandwidth, reliable network connectivity and support for Artificial Intelligence (AI) based applications. This paper investigates the use of solar powered micro edge data centers in NextG RANs, which will reduce the carbon emissions of network operators and can ensure continuous operation during power outages. We adopt a nanogrid power architecture that integrates solar power generation and energy storage co-location with small edge data centers. A novel methodology is proposed that uses data driven techniques to identify the power outage prone sections of a city and the solar generation potential of rooftops in that part of the city. Given a set of rooftops that meet or exceed the edge data center energy requirements, unsupervised machine learning is used to determine the geographic locations to place micro edge data centers. We illustrate our methodology using New Your City as a case study. This work provides insights for urban planners, network operators, and policymakers aiming to develop sustainable and resilient NextG RANs for societal good.