A gamma autoregressive video model on ATM networks

Several works have concluded that the number of cells per frame of variable bit-rate (VBR) video teleconference follows a gamma distribution. But few video models meet this characteristics. In this paper we propose a novel video model, called first-order autoregressive (AR) gamma sequence, in which random variables are marginally distributed as gamma variables. We give the definition of a GAR model and a simple realization method of GAR(1). Parameter estimation methods are studied, too. Simulation results show that our model can meet the characteristics of VBR video better than existing models.