Performance Evaluation of Neural Network-based Offset Optimization in Cell Range Expansion for Multiple Frequency Bands

The amount of data traffic over the mobile network is still increasing. To support such large amount of traffic, additional higher frequency usage is one of the promising techniques. The network needs to carefully associate the users to the base stations and frequency bands, since users can connect multiple BSs and multiple frequency bands in such deployments. In order to offload the users effectively, the cell range expansion (CRE) is effective, and the offset values can be optimized by the neural network (NN). In the paper, we evaluate the performance of the NN-based offset optimization in CRE for multiple frequency bands. Simulation results show the good trade-off between the performance and the amount of feedback from the users.