Modeling aeronautical data traffic demand

Determining the amount of air traffic in an area can help to identify the traffic demands of current and future airborne communication systems. Many mobility models for pedestrians and road vehicles exist, but hardly any for aircraft. Aircraft traces for three exemplary scenarios are analyzed to derive such a model with regard to traffic demand in a certain area, e.g. a cell. The number of aircraft inside the area is considered as state space of a Markov Process. For each state the arrival and departure process is fitted according to the hypothesis to follow a Poisson distribution. The goodness of fit is evaluated by Chi-Squared testing. Results considering all aircraft show predominant Poissonian behavior for areas with low aircraft density and little take-off and landing activities due to airports. In an urban area with high aircraft density and several airports the Chi-Squared Test often rejects the hypotheses. In a second step the problem was limited to only analyze transit traffic since take-off and landing seem to follow a different statistical behavior. The number of states where the Chi-Squared Test passes increases significantly for the scenarios with high aircraft densities. This work shows that aircraft arrivals and departures can be modeled as a Markov Process with state-dependent arrival and departure rates λ(i) and μ(i).