Graph Coloring and Ant Colony Optimization based Sub-channel Allocation Algorithms for LTE-A HetNets

LTE-Advanced is a promising technology which supports much higher peak rates, higher throughput and coverage, and lower latencies, resulting in a better user experience. For ex-tended indoor coverage LTE-A supports two-tier network composed of conventional macro-cellular networks and femtocell networks known as HetNets. As the femtocell shares the same frequency band with underlying macrocell, the cross tier interference needs to be mitigated. The inter-femtocell and cross tier interference near femtocell boundary may result in unwarranted degradation of system performance. In LTE-A, as an OFDM based technology, we consider the use of prudent PRB (Physical Resource Block) allocation to mitigate downlink inter-femtocell interference as well as cross tier interference in the mentioned environment for this small HetNet. Allocation of PRBs to network users is formulated as a graph coloring problem. Based on this interpretation and initial sensitivity study, we propose dynamic resource allocation algorithms namely greedy allocation (GA), SINR based allocation (SINRA) and ant colony allocation (ACA). From the simulation results and its analysis we can conclude that, ACA algorithm outperforms rest of the considered algorithms while GA and SINRA provide performance with the significant improvement over random allocation (RA) algorithm as far as performance parameters such as averaged SINR experienced by a network user, outage probability and number of PRBs needed are concerned.