|Authors||O. Lysne, A. Elmokashfi, N. N. Schia, L. Gjesvik and K. Friis|
|Title||Critical communication infrastructures and Huawei|
|Project(s)||The Center for Resilient Networks and Applications|
|Publication Type||Proceedings, refereed|
|Year of Publication||2019|
|Conference Name||TPRC48: Research Conference on Communications, Information and Internet Policy|
Recently, there have been growing cyber-safety concerns over telecom equipment made by the Chinese vendor Huawei. This has led many countries to ban Huawei from supplying equipment for building the next generation of mobile networks, 5G. Responses from mobile operators and the telecom community in general have been mixed. For instance, many European mobile operators have stated that these concerns are overblown and that such a ban would delay 5G rollout by two to three years in the best case. Moreover, some operators have directly questioned the ability of the other vendors to timely deliver a complete 5G network. However, these claims have mostly not been grounded in empirical data. This paper takes a multi-perspective approach to investigating this problem empirically. We start by categorizing responses from different countries to using Huawei equipment in 5G. We then analyze the importance and readiness of Huawei for supplying 5G equipment. This analysis is based on contributions to standards and patents. We also present a conceptual risk analysis framework to qualitatively evaluate the ability of a single vendor to cause considerable damage to critical communication infrastructures. This model aims at exploring a set of relevant axis. More specifically, we look at potential for harm in different political climates that is peace, crisis and war. Another axis is whether banning a particular vendor from supplying equipment for the upcoming mobile networks generation is useful without having a backward compatible ban. A third axis is the ability of a vendor to cause harm as a function of the type of supplied equipment, for example radio towers vs network management systems. Combining the analysis of readiness for supplying 5G and potential for causing harm allows us to roughly estimate the likely impact that a complete ban would have on 5G rollout in different parts of the world. We find that such a ban can possibly delay 5G by two years or more for operators with high dependence on Huawei. Consequently, we explore potential approaches that would both reduce vendor-related risk and do not significantly delay the rollout of 5G. These include heterogeneous multi-vendor deployments, equipment verification and testing, international collaboration as well as signing non-aggression treaties. Unfortunately, there is no technological solution that fully remedy this problem. Combining technical solutions with efforts to build trust between countries, enforce existing alignments or create new ones seems a promising way forward.