Published June 27, 2014
| Version v1
Conference paper
Open
Optimization of Cloud-RAN deployment using Integer Programming
Authors/Creators
- 1. DTU Department of Photonics Engineering Technical University of Denmark Denmark
Description
The mobile data traffic is increasing with a very high rate due to the rise of smartphones and
tables. The only way to overcome such a large traffic volume is to increase the capacity in
the network by deploying more base stations. This means, that the mobile operators are
facing a challenging future as the cost to build, operate and upgrade the Radio Access
Network (RAN) is becoming more expensive. Due to a very competitive marked, the
Average Revenue Per User (ARPU) is constant or even decreasing. It is therefore in the
mobile operators' interest to find new solutions on how to reduce cost in order to maintain a
profitably business.
It is estimated that Information and Communication Technologies (ICT) accounts for 2-4% of
the worldwide carbon emissions and is expected to double in 2020. The mobile network has
a big part in this and is therefore relevant when talking Green Energy. In the mobile network,
base stations accounts for 58% of the power consumption. This means that the increasing
number of base stations in the network will be a major problem when trying to lower carbon
emission in the future.
A promising solution to the above challenges is C-RAN (Cloud-Radio Access Network),
which is a novel mobile network architecture. C-RAN is using distributed Remote Radio
Head (RRH) and centralized Baseband Unit (BBU) architecture. C-RAN has the potential to
support extremely dense mobile network deployments enhancing the network capacity while
offering cost savings and lower power consumption on baseband resources.
This project will look into solutions on how the operators can optimize new green field
deployments in terms of Total Cost of Ownership (TCO). In order to minimize the TCO, an
Integer Programming (ILP) model is made. The model is optimizing the allocation of RRHs
to the BBU pools, when looking at different mobile network deployment scenarios. The
model takes into account that all cells are having different traffic profiles and that different
types of base stations can be used.
By using this model, the power consumption in the mobile network will greatly decrease, as
fewer base stations will be needed compared to a traditional mobile network deployment.