Cooperative and Adaptive Load-Aware Handover Optimization in Heterogeneous Networks

In heterogeneous networks (HetNets), the deployment of small base stations (SBSs) to offload traffic from macro base stations (MBSs) may unintentionally increase signal interference due to their dense deployment and transmit power (TXP) disparities among base stations (BSs). This interference significantly degrades the signal quality received by the user equipment (UE), resulting in frequent handovers and unstable connectivity. To address this issue, we propose a cooperative and adaptive load-aware mobility robustness optimization (MRO) scheme. In this scheme, agents assigned to MBSs and SBSs jointly optimize TXP and handover control parameters using a multi-objective learning approach. To strike a balance between connection stability and load balance, agents explore diverse solutions and adaptively select the most suitable one based on current network conditions. In addition, a team reward-based feedback mechanism is incorporated to facilitate cooperative learning and align the actions of local agents with global network goals. The simulation results confirm that the proposed scheme mitigates connectivity degradation while effectively reducing the average MBS load by 19.6% and decreasing the number of handovers, ping-pong handovers, and radio link failures throughout the network by 25.0%, 41.9%, and 38.2%, respectively.