The recruiting process has already been carried out by the University of Seville, and the 3 engineers have been hired. In addition, the architecture of the Cloud Manager for accessing the remote laboratory that will be installed at ETSI by summer 2019 has been reworked. The first optimization modules in which the members of the WiFiciency project will work will be:
– Automatic interference classifier for WiFi (external networks) and non WiFi (Bluetooth, microwave oven, jammer, etc) signals. It is proposed to use 1D convolutional neural networks to implement a classifier.
– Proactive Load Balancing (PLB) to transfer devices between radios and between APs before congestion occurs. It will use the radio channel usage metric.
– Adaptive Traffic Control (A-TC) to dynamically limit network traffic if necessary, avoiding saturation. For this purpose, the reinforcement learning paradigm will be used.
These algorithms are in the design phase and if the simulations are validated, they will move on to the development phase over the next few months. We are also working on an article ( from the study of the state of the art thanks to the WiFiciency project) to be presented at the URSI 2019 congress to be held in Seville in September.
Project WiFiciency is supported by: