A general performance model for mobile slotted ALOHA networks with capture

A general analytical model is proposed to evaluate the performance of a mobile slotted ALOHA network. Near-far effects are incorporated in the model (in an approximate way) by considering a multi-group system. The model approximates the network behavior by an iterative solution between multiple one dimensional Markov chains, one for each group. The model is quite broadly applicable since (1) different physical layer parameters, such as modulation and coding, are abstracted by a set of conditional success probabilities, and (2) different groups can have different link layer parameters, such as the probabilities of packet generation and retransmission. As an example, a 5-group system with fading is used to illustrate the usefulness of the proposed model. The agreement between the analytical and simulation results is good. It is found that the impact of the fading effect on network performance depends on the significance of the near-far effect. The problem of unbalanced throughput between near and far groups is solved by assigning different retransmission probabilities to different groups.