Deep Learning Based CoMP Transmission Method Using Vehicle Position Information for Taxi Radio Systems

This paper proposes a coordinated multi-point (CoMP) transmission method based on deep learning for taxi radio systems to prevent inter-cell interference (ICI). In CoMP, it is essential to select whether to use simultaneous transmission or time division multiplexing (TDM) considering the effect of the ICI. The feature of the proposed method is to determine such a transmission mode by using vehicle position information as taxi radio systems have such position information. Moreover, the proposed method makes it possible to avoid the online system capacity calculation also required for CoMP thanks to the use of deep learning. The effectiveness of the proposed method is demonstrated in comparison with a traditional online calculation method under the practical scenario based on the taxi radio system in Japan.