QoS Oriented Traffic Modeling of Mobile POS Service in Cellular network

The market potential of mobile point of sale (POS) service is attractive to the mobile network operators, while the traffic challenges caused by heavy traffic is becoming a critical issue to the operators. In order to guarantee the Quality of Service (QoS) and make best use of the existing network, it is important to model the traffic of mobile POS service. A tree-structured modeling method is presented in the paper, which can be used to model the traffic of mobile POS service as well as other interactive M2M services, such as remote medical services. The modeling method is a combination of the layered modeling method used in traditional H2H communications and the transaction tree modeling method used in bank trade. As another main point, a specific traffic model of mobile POS service is given in the paper. There are five parameters in the traffic model, including request packet size, response packet size, the interval between request and response, the interval between two requests, and the transmission probability. After counting the packet length of each message, calculating the average length, analyzing the effect of difference between the actual and average length on the cellular network, we find that all the request type of mobile PUS service could be simplified as the average length. Further on, a curve-fitting method is used to obtain the distribution of the intervals. All the data used in the modeling are collected from the traffic of real network. The QoS attributes assumption of mobile PUS service including the maximum accessing user number, the packet loss rate and the end-to-end delay is given in the article and is validated by the simulation using the specific model of mobile PUS service provided in the paper. The assumption is consistent with the simulation result, and is proved to be reasonable. The results can be used as a reference when implementing mobile POS service in cellular network.