Analysis of a Transportation Problem based scheduling algorithm for inter-band carrier aggregation

Carrier aggregation is one of the new features of LTE-Advanced in order to provide higher user throughputs. Scheduling algorithms must guarantee backward compatibility with Release 8 users, so that they must be assigned to certain a Component Carrier (CC). This paper formulates this assignment as a Transportation Problem (TP) where the costs of shipping from a supply point to a demand point are oppositely related to the spectral efficiency. The benefits of this algorithm for inter-band spectrum aggregation are shown here. Its performance has been analyzed for aggregation of CCs at frequency bands with very different propagation characteristics, i.e. 800 MHz and 2.6 GHz, which are current candidates for LTE-A deployments in Europe. Simulation results show that this algorithm can provide up to 50% gains over a random scheduler for low user densities and large cell radii, which makes it especially appropriate for rural environments. On the other hand, the fairness, cell capacity and gain on cell-edge user throughput increase as the number of LTE-Advanced users increases.