GA Based Multi-Objective LTE Scheduler

As the complexity of networks increases significantly, cognitive networking becomes an essential tool to provide efficient management of the valuable resources. In this paper, a Genetic Algorithm (GA) based cognitive LTE downlink scheduler is proposed to allocate radio resources to the users. In the proposed scheme, a network administrator defines high level network policies by setting the operational mode to throughput or fairness and a target threshold for the selected mode. For example, the proposed scheduler dynamically and quickly adapts its decisions to ensure the best fairness among the solutions satisfying the target throughput or the highest throughput if none of the solutions achieves the desired throughput. We implemented a C# based simulation tool and demonstrated the trade-off between the convergence speed and the quality of the solution by varying the parameters of LTE and GA. Numerical results demonstrated that the proposed GA scheduler can be effectively used to manage throughput and fairness objectives in dynamic network scenarios.