Creating and sharing knowledge for telecommunications

A flexible physical layer and fronthaul research testbed for C-RAN

Santos, J. ; Dinis, DACD ; Riscado, D. ; Anjos, G. ; Belo, D. ; Oliveira, A. ; Monteiro, P. ; Carvalho, N.B.C.

Microprocessors and Microsystems Vol. 00, Nº 00, pp. 1 - 11, September, 2016.

ISSN (print): 0141-9331
ISSN (online):

Journal Impact Factor: (in )

Digital Object Identifier: 10.1016/j.micpro.2016.09.001

Abstract
Mobile networks are subject to an explosive increase in data traffic, in a context of continuous mobility
and more stringent levels of QoS, which imposes demanding requirements to telecommunication networks.
To cope with this trend, a novel paradigm of radio access networks, known as C-RAN, is being
developed, where the physical layer processing is also shifted from the edges of the network to a centralized
location. C-RAN provides important benefits and will be one of the cornerstones of 5G communication
systems. However, some architectural and implementation tradeoffs need to be further evaluated.
Moreover, the modularity and extensibility of research platforms supporting C-RAN is still very restrictive.
This paper presents a laboratorial platform aimed for the development and trial of C-RAN compliant
features. The proposed testbed is very modular and flexible and it is intended to provide a cost-effective
emulation and physical layer implementation platform for the main C-RAN modules, namely the BBU, the
fronthaul and the RRHs. Based on open FPGA platforms, it features a high level of flexibility in terms of
configurations, waveforms and interfaces, and includes all the components required to build an open and
complete C-RAN compliant base station. It is mainly used for the experimentation and evaluation of next
generation wireless communication systems, including new fronthaul protocols and interfaces as well as
5G waveforms. It integrates a 25 km optical fronthaul, a software defined multi-mode and multi-band
RF front-end and a digital radio compression algorithm associated with the optical fronthaul. The inclusion
of low-latency (de)compression algorithms was of paramount importance in order to achieve a 50%
reduction in terms of fronthaul bandwidth.