Performance Analysis of OpenAirInterface System Emulation

With the rapid growth of mobile networks, the radio access network becomes more and more costly to deploy, operate, maintain and upgrade. The most effective answer to this problem lies in the centralization and virtualization of the eNodeBs. This solution is known as Cloud RAN and is one of the key topics in the development of fifth generation networks. Within this context OpenAirInterface is a promising emulation tool that can be used for prototyping innovative scheduling algorithms, making the most of the new architecture. In this work we first describe the emulation environment of OpenAirInterface and its scheduling framework and we use it to implement two MAC schedulers. Moreover we validate the above schedulers and we perform a thorough profiling of OpenAirInterface, in terms of both memory occupancy and execution time. Our results show that OpenAirInterface can be effectively used for prototyping scheduling algorithms in emulated LTE networks.