Litcius/Paper detail

The Cost of Securing O-RAN

Joshua Groen, Brian Kim, Kaushik Chowdhury

202313 citationsDOI

Abstract

A promising vision for the emerging next generation of cellular networks is one that embraces openness, intelligence, virtualization, and distributed computing. The Open Radio Access Network (O-RAN) framework is making significant strides toward these goals and is already seeing prototype deployments in academia and industry. While there is general consensus that this technology may disrupt the status quo by eliminating vendor lock-ins, there are serious questions about the security implications in such dis-aggregated networks. Indeed, securing data and controlling interfaces must be a core consideration in the design of O-RAN and cost/benefit tradeoffs need to be rigorously analyzed, given the short time-scales of wireless operation. In this paper, we undertake the first systematic study on the impact of encryption on a critical O-RAN interface (called ‘E2’) connecting the base station to a near-real time radio intelligence controller using an implementation on the Colosseum radio frequency (RF) emulator. The contributions of this paper include quantitative measurements of added latency and CPU utilization due to encryption on the E2 interface that could impact data acquisition and machine learning models. In our experiments we found encryption adds <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\leq 50\mu s$</tex> of delay and CPU utilization limits throughput to approximately 500 Mbps. We also include a theoretical model to extend this study to other O-RAN implementations beyond the emulation environments of the Colosseum.

Topics & Concepts

Computer scienceEmulationEncryptionRadio access networkVirtualizationBase stationC-RANInterface (matter)RanCloud computingOperating systemComputer networkEconomic growthBubbleMobile stationMaximum bubble pressure methodEconomicsCooperative Communication and Network CodingAdvanced MIMO Systems OptimizationOpportunistic and Delay-Tolerant Networks