Using MDT data for 5G initial access optimisation

This work arises from the desire to reduce latency during the initial access phase in 5G networks. Millimeter wave (mmWave) bands are a key technology since they provide a great availability of bandwidth, essential to deliver the continuously increasing demand of data without saturating the conventional bands (below 6 GHz). Despite the great advantages this band can offer, at these frequencies the propagation characteristics, mainly in terms of attenuation and penetration losses, are significantly different from those in the conventional bands, thereby highly directional transmissions are crucial. The main goal of this thesis is to identify an intelligent sweeping pattern making use of contextual information extracted from MDT (Minimization of Drive Test) data provided by Telecom Italia, which enable a collection of user equipment (UE) radio measurements combined with GPS coordinates. MDT data represents a fundamental tool to create a historical database containing information regarding the most populated area in terms of data traffic, at different times of the day. Else, these data are employed to develop a dual-slope model accounting for the propagation effects in each cell of Corticella, a fraction of the municipality of Bologna: this allows to plainer define the reference environment over which the initial access optimization algorithm applies. Afterwards, the historical database is used to choose the sequence of beams to be followed by the base station (gNB) and it is constantly updated according to the new collected information.