RACH Optimization with Decision Tree Based Supervised Learning for Conditional Handover in 5G Beamformed Systems

Abstract: Higher frequencies that are introduced in 5G networks cause rapid signal degradation and challenge user mobility. In recent studies, a conditional handover procedure has been adopted for 5G networks to enhance user mobility robustness. In this paper, mobility performance of the conditional handover is analysed for 5G mm-Wave systems with beamforming. In addition, a resource efficient random access procedure is proposed that increases the chance of contention-free random access during handover, which reduces signaling and interruption time. Moreover, simple, yet, effective decision tree based supervised learning method is proposed to minimize the handover failures that are caused by beam preparation phase of random access procedure. Results reveal the trade-off between contention free random access and handover failures. It is also shown that the optimum operation point of random access is achievable with proposed learning algorithm for conditional handover.