The centralized processing in cloud radio access network enables cooperation between baseband processing units (BBUs)\nlike inter-cell interference (ICI) cancellation on the basis of coordinated multipoint (CoMP). Large amounts of the sharing\ndata will be transmitted through fronthaul transport network. In the paper, both integer non-linear programming (INLP)\noptimization model and adaptive genetic algorithm (GA) are explored to release the capacity pressure of the fronthaul\ntransport network when CoMP is introduced. We also consider the resource allocation problem of the passive optical\nfronthaul network. The proposed algorithm tries to reduce the downlink bandwidth and improve the optical resource\nallocation efficiency of the optical fronthaul with minimal influence on the fronthaul topology. During the simulations,\nthree critical factors are considered: (1) the number of cell edge users, (2) the average traffic demand of cell edge users,\n(3) the size of cell cluster used to enable the CoMP. The simulation results show that the most efficient bandwidth saving\nand optical resource allocation can be achieved with INLP, while the proposed adaptive GA nearly has the same\nperformance with low computational complexity and fast convergence, which is more applicable for the large-scale\nfronthaul network. Furthermore, the load difference of the fronthaul transport network can be further reduced.
Loading....