Optimal fronthaul capacity allocation and training for joint uplink receiver in C-RAN

Dora Boviz (1) , Sheng Yang (2)

1 Nokia Bell Labs [Nozay]
2 L2S - Laboratoire des signaux et systèmes

In Cloud Radio Access Networks (C-RAN) one of the main challenges is accurate data transfer on limited capacity fronthaul links between Remote Radio Heads (RRHs) and the central processing unit. Particularly in large-band and multi-antenna transmissions, fronthaul capacity can be a limiting factor. In this paper we study the impact of non-ideal fronthaul on uplink received signal, channel estimates and consequently on the achievable rates. We consider both the fronthaul capacity and the length of training sequences variable and find optimal values allowing high-rate transmission taking into account the costs of over-the-air transmission and fronthaul usage, thus, our approach can be used in dynamic system design and configuration given technical and economical factors.

Domains: Networking and Internet Architecture [cs.NI]; Information Theory [cs.IT]

Dora Boviz, Sheng Yang. Optimal fronthaul capacity allocation and training for joint uplink receiver in C-RAN. European Wireless Conference, May 2016, Oulu, Finland. ⟨10.1109/iraniancee.2017.7985380⟩. ⟨hal-01320512⟩