| Multi-connectivity, which allows a user equipment to be simultaneously connected to multiple cells from different radio access network nodes that can be from a single or multiple radio access technologies, has emerged as a useful feature to handle the traffic in heterogeneous cellular scenarios and fulfill high data rate and reliability requirements. This paper proposes the use of deep reinforcement learning to optimally split the traffic among cells when multi-connectivity is considered in a heterogeneous 4G/5G networks scenario. Obtained results reveal a promising capability of the proposed Deep Q Network solution to select quasi optimum traffic splits depending on the current traffic and radio conditions in the considered scenario. Moreover, the paper analyses the robustness of the obtained policy in front of variations with respect to the conditions used during the training. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.