Mathematical analysis for corrective performance evaluation of call based QoS in active measurement

In this paper, we addressed the problem of improving call-based QoS measurement accuracy with the focus on call blocking probability as monitored via the active measurement or test drives in cellular-type networks. The system has been modelled as continuous time Markov chains with necessary assumption of Poisson traffics for both actual load and injected or intrusive load. With a product-form solution, a convenient formula has been derived to map the value of call blocking probability as seen by the intrusive traffic to the true value of call blocking probability of actual traffic in the non-intrusive system. With the derived corrective factor, QoS measurement can be accurately measured and regulated. Consequently, the cellular-type networks so widely used for wide-area, mobile communication platforms of ITS can be better engineered and hopefully become a dependable foundation for ongoing as well as future ITS developments.