Handover Prediction for Aircraft Dual Connectivity Using Model Predictive Control

Abstract

Providing connectivity to aircraft such as flying taxis is a significant challenge for tomorrow's aviation communication systems. One major problem is to provide ground to air (G2A) connectivity, especially in the airport, rural and sub-rural areas where the number of radio ground stations is not adequate to support the data link resulting in frequent interruption. Hence, effective handover decision-making is necessary to provide uninterrupted services to aircraft while moving from one domain to another. However, the existing handover decision is not efficient enough to solve the aircraft connectivity in such airspace. To overcome this problem, a prediction based optimal solution to handover decision making (handover prediction) would be appropriate to provide seamless dual connectivity to aircraft. In this paper, the handover prediction problem is formulated as a constrained optimization problem in the framework of the model for predictive control (MPC). The cost function and the constraints are derived in terms of dual connectivity variables over the prediction horizon. This problem is solved using a two-dimensional genetic algorithm (2D-GA) to obtain the predictive optimal handover solution. Simulation results show that the proposed dual connectivity handover can significantly improve the handover success probability. Finally, our results show that network densification and predictive control model have improved aircraft performance.