Distributed and Adaptive Optimization of LTE-TDD Configuration Based on UE Traffic Type

Interference mitigation is one of the main challenges in LTE TDD networks, which requires that all cells in the network follow the same TDD UL/DL configuration. However, different UEs in the network may have different UL or DL traffic types, which may also vary with time; as a result, different cells may require different TDD UL/DL configurations in time. In this paper, we consider an LTE network, where each eNodeB in the network has backhaul links, i.e. X2 interfaces, only with its immediate neighboring eNodeBs. We investigate an adaptive optimization of TDD UL/DL configuration index in mean square error (MSE) sense based on the available UL and DL traffic of all RRC connected UEs in different cells within the network. We then propose a distributed algorithm, where all eNodeBs in the network will reach consensus on the optimum TDD UL/DL configuration through local communications with their neighboring eNodeBs over X2 interfaces. Our simulation results show that the LTE network, which is running the proposed algorithm, provides the better utilization of the available frequency-time resources and achieves a higher UE Throughput compared to a fixed TDD UL/DL configuration.