The following contribution has been supported by the EU project MEDIEVAL .
By cloning this repo you can access the first draft of our error model for LTE.
The basic module used for this work is the one released by Giuseppe Piro during the GSOC '10 ,
but our final contribution has already been adapted to the latest version of the LTE-EPC Network Simulator (LENA) project .
In the context of LTE, link abstraction models are needed to condense the wideband channel quality measurements performed by the User Equipment (UE) into a small set of Channel Quality Indicators (CQI). To estimate the system performance, a link performance model is then required to produce the results for evaluating the link quality, which is based on resource and power allocation in a certain radio propagation environment. With the proposed modelling technique, accurate link abstraction can be obtained based on the mean mutual information per coded bit (MMIB) metric which is the mean mutual information between coded bits and their Log- Likelihood Ratio (LLR) values. Then, a mapping function providing the associated Block Error Rate (BLER), that represents the Link to System Level (L2S) interface, is integrated in ns-3 in order to design an efficient system error model: a binary random variable with that probability is then drawn and a corresponding error event is generated.
To do so, we mathematically
describe the BLER via three parameters: the mutual information per coded
bit (MIB), b and c (that depend on the selected MCS) . In order to get MCS
dependant parameters, we performed a fitting based on the curves generated
with a Link Level simulator . The reliability of such approach is
provided through graphics and matlab codes that you can find in
You can visualize the functioning of this contribution by simply running
the example /src/devices/lte/examples/lte-errormodel.cc.
So far, the main contribution can be found in:
Anyway, we are planning to further contribute by developing an efficient
resource allocator, and by introducing LTE HARQ techniques (CC-IRv1-IRv2).
 IEEE 802.16m Evaluation Methodology Document (EMD), **2008-07-03*