Collaborative Research: NeTS: Medium: A Systematic Approach to Diagnostic Testing of Next-Generation Radio Access Networks

The radio access network (RAN) is an important subsystem of 5G networks and beyond – providing universal coverage and ubiquitous Internet access to billions of mobile users. Open radio access network (O-RAN) promotes an open system approach, where components and products from different vendors can interoperate with each other, thus accelerating technology innovations. This thus calls for proper testing which has not received its due attention from the research community. This project seeks to identify and address this new research topic of mobile network diagnostic testing, particularly for the O-RAN subsystem. New diagnostic testing methods are developed to not only determine whether the tests pass or fail, but also diagnose the root causes to learn why. These new designs and their gained insights lead to improvements in operation correctness, interoperability, performance and security of the upcoming RAN infrastructure and jumpstart the growth of O-RAN and future mobile technology ecosystem. The project also seeks to influence the standardization of new 5G releases and 5G beyond technologies. It recruits fresh talents and train a new generation of students and engineers for future mobile Internet design. This project explores a systematic approach to mobile network diagnostic testing, spanning from fresh-view modeling and abstraction, efficient algorithms, to novel instruments for tracing and diagnosis, to address all limitations of RAN testing. The research tasks enhance RAN diagnostic testing in three aspects: (1) near-complete test coverage by exploiting dependency among multiple procedures and multiple interfaces of RAN software to generate missing test cases and pinpoint their root causes, (2) non-intrusive, end-to-end tests to assess full-stack performance and potential threats of applications on commodity 5G devices, and (3) test efficiency to optimize test runs and reuse prior test results by identifying repetitive operations and assessing the updates over evolving RAN technology releases. The proposed testing designs are integrated to an open-source toolkit and evaluated in the lab testbed and real-world O-RAN systems with industry collaborators to facilitate technology transfer. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.