Traffic modelling for mobile wireless applications

Classical fixed wireline telephone traffic prediction theory is based on temporal call arrival and call service probabilities, usually exponential. For such systems, the grade of service that can be provided is dependent on the number of available fixed wireline phone sets and the expected call traffic. With the introduction into the market place of mobile commercial cellular and indoor wireless PBX based phone systems, fixed wireline sets are replaced by base stations and portable phone sets. The amount of traffic that a base station will have to accommodate is a function of the number of portables that need to be serviced by the base station and the traffic per portable. The traffic per portable can be analyzed using classical traffic analysis methods. However little analysis is available to account for the overall grade-of-service that can be provided by a cellular based architecture, as a function of the spatial distribution of the portables in the system. The paper proposes analytic models to predict traffic capacity for mobile systems based on both the temporal call arrival statistics as well as the spatial distribution of the portables in the system. The grade-of-service is calculated as a function of the spatial distribution and the call arrival statistics.