Chance-Constrained Spectrum Allocation for Fair LTE-U/Wi-Fi Coexistence

In order to meet the ever-increasing data rate demands of future mobile networks, LTE in the unlicensed band (LTE-U) has been proposed in which LTE is operated on the unlicensed spectrum along with the incumbent unlicensed spectrum technologies e.g. Wi-Fi However, enabling LTE-U with the Wi-Fi network in the same unlicensed spectrum results in severe Wi-Fi performance degradations. This paper addressed such coexistence problem of Wi-Fi and LTE-U, by formulating a chance-constrained optimization problem such as to maintain certain Wi-Fi performance. In the first step, unlicensed spectrum is shared among Wi-Fi and LTE-U by adjusting the so-called duty-cycle according to the random Wi-Fi performance. Secondly, the LTE-U user association is performed through knapsack problem to efficiently utilize the available LTE-U duty-cycle. Simulations results show that the proposed algorithm provides fairness to the Wi-Fi system by adjusting the LTE-U duty-cycle.