This paper investigates moving networks of Unmanned Aerial Vehicles to extend connectivity and guarantee data rates in the 5G by analysing possible hovering locations based on limitations such as flight time and coverage. The authors provide analytic bounds on the requirements in terms of connectivity extension for vehicular networks served by fixed Enhanced Mobile BroadBand infrastructure, where both vehicular networks and infrastructures are modelled using stochastic and fractal geometry as a model for urban environment. The authors prove that assuming n mobile nodes (distributed according to a hyperfractal distribution of dimension d F) and an average of ρ Next Generation NodeB (gNBs), distributed like a hyperfractal of dimension d r if ρ = n θ with θ > d r /4 and letting n tending to infinity (to reflect megalopolis cities), then the average fraction of mobile nodes not covered by a gNB tends to zero like O n − d F −2 ð Þ dr 2θ − dr 2 ð Þ. Interestingly, the authors prove that the average number of drones, needed to connect each mobile node not covered by gNBs, is comparable to the number of isolated mobile nodes. The authors complete the characterisation by proving that when θ < d r /4 the proportion of covered mobile nodes tends to zero. The authors offer insights on the placement of the 'garage of drones', the home location of these nomadic infrastructure nodes, to minimise the 'flight-to-coverage time'. The authors provide a fast procedure to select the relays that will be garages (and store drones) in order to minimise the number of garages and minimise the delay. The authors' analytical results are confirmed using simulations in Matlab.