SUMMARY REMARKS & TESTBEDS
In wireless networking R&D we typically depend on experimentation to make realistic evaluations, as simulation is inherently a simplification of the real-world. However, experimentation is limited in aspects where simulation excels, such as repeatability and reproducibility.
Wireless experiments are hardly repeatable. Given the same input they can produce very different output results, since they are influenced by external random phenomena such as noise, interference, and multipath. Experiments are also difficult to reproduce: either the original testbed is unavailable or becomes inaccessible.
Fed4FIRE+ wireless testbeds such as w-iLab.t and CityLab do not fully address the problem. The CONCRETE tool used in such testbeds targets repeatability and reproducibility of experiments but ignores all unstable executions, which are also representative of the system operation.
What if we could make any wireless experiment repeatable and reproducible under the same exact conditions? What if we could share the same Fed4FIRE+ testbed execution conditions among an “infinite” number of users? What if we could run experiments in faster than real time?
INESC TEC has developed an approach that combines the best of simulation and experimentation to enable exactly that, towards the concept of Testbed-In-a-Box in complement to testbeds. By relying on ns-3 and its accurate simulation of the MAC and upper layers, we have been exploring how ns-3 can be used to replicate wireless experiments in a “box”. The SIMBED OC3 project validated our approach for controlled environments and helped identifying some of its limitations.
The SIMBED+ project aims to continue this work by running a new set of wireless experiments for further validating an improved version of our approach in non-controlled environment. For that purpose, we will leverage our custom ns-3 model, and the Fed4FIRE+ w-iLab.t and CityLab testbeds enabling indoor and outdoor wireless experiments with the most recent Wi-Fi variants.