Dynamic packet assignment in OFDM based cellular systems - performance analysis

In this paper, we present a performance analysis of OFDM based cellular systems when using dynamic packet assignment in the uplink. We propose a two dimensional Markov chain to model the state of the traffic in each cell of a cellular system with data-only traffic and solve this Markov chain to compute the buffering probability, mean delay, and the average system throughput. We perform simulations to determine the optimum position of the base station at which we obtain the best system performance in terms of mean delay. We show that dynamic packet assignment provides improved performance compared to dynamic channel assignment in terms of Erlang capacity. We also show that our analytical model is accurate and computationally efficient.