Cross-layer based optimization for fairness control mechanism in OFDM wireless networks

The important problem of resource allocation in multi-user wireless networks is that to achieve the balance between efficience and fairness. In this paper, by using the cross-layer optimization mechanism, we derive an optimization model to obtain efficient and fair downlink sub-carrier allocation in multi-user OFDM wireless networks. A dynamic sub-carrier allocation algorithm (DSAA) is proposed in order to obtain the maximization of the total scheduling utility based on the optimization model and the maximization utility theory. It is shown that by combining DSAA with time scale interference predictor (TSIP) which at large time scales predict ON / OFF period of user data with temporal correlation structure across multiple time scales in multi-user interference environment, efficiency is improved further. Preliminary simulation results verify the efficiency and fairness of the scheme.