A queuing delay model for dimensioning interactive service traffic in HSDPA networks

This paper considers interactive traffic queuing delay estimation in high speed downlink packet access (HSDPA) networks. The Web browsing is the most widely used interactive application. Its traffic model is subdivided into three levels: session level, burst level and packet level. Based on this model, the statistical behavior of the queueing delay is studied. It is shown that the queueing delay for the Web traffic follows an exponential distribution. Analytical modelling of the probability density function (PDF) of the queueing delay being untractable, we resort to simulated data and provide simple mathematical formulation of the different parameters which characterizes the density function. Indeed, we present useful equations which could be used, directly, in HSDPA network dimensioning, and as a reference, to satisfy a certain quality of service.