Research On Interference Coexistence Technologies Of Cellular Networks In Unlicensed Spectrum

With the increase of mobile data traffic and intelligent terminals in the future, the current limited spectrum resources is difficult to meet the requirements. Therefore, it is imperative to add new spectrum resources.Cellular networks in unlicensed spectrum, which utilize carrier aggregation to put licensed spectrum and unlicensed spectrum together to provide greater system capacity and faster transmission rates. However,due to the openness of unlicensed spectrum, cellular networks in unlicensed spectrum must achieve the coexistence with other radio access networks. Therefore, it is necessary to study interference coexistence technology of cellular networks in unlicensed spectrum.This topic comes from the horizontal project "research on architecture evolution strategy of broadband mobile communication future network". According to the design requirements of different countries and regions, cellular networks in unlicensed spectrum include Licensed-Assisted Access using LTE (LAA-LTE) and LTE in unlicensed spectrum (LTE-U), which ensure the deployment of cellular networks in unlicensed spectrum to a certain extent. However, the existing cellular networks in unlicensed spectrum exist issues including can not effectively deal with network traffic load, high interference on other wireless access networks in unlicensed spectrum , and so on. In order to deal with the above-mentioned problems, this paper includes:Firstly, this paper introduces the current development of cellular networks in unlicensed spectrum, and summarizes the interference coexistence technologies of cellular networks in unlicensed spectrum,including Listen-Before-Talk, channel selection and blank subframe allocation.Then,an adaptive threshold LBT algorithm is proposed for not being effectively deal with traffic load in existing LBT algorithm in LAA-LTE system. The algorithm considers both traffic load of LAA-LTE base station and user QoS guarantee, and dynamically adjusts the clear channel accessment threshold based on the traffic load of the base station, and changes the transmission strategy of the base station, thus improving the efficiency of the LAA-LTE system. The simulation results show that the adaptive threshold LBT algorithm proposed in this paper greatly improves the performance of LAA-LTE at the expense of acceptable performance of Wi-Fi, especially in the heavy traffic load conditions of LAA-LTE base station.Finally, a dynamic subframe allocation algorithm based on reinforcement learning is proposed for low transmission efficiency in existing subframe allocation algorithm in LTE-U system. The algorithm dynamically adjusts the transmission power level and blank subframe allocation of LTE-U base stations by Q learning algorithm in reinforcement learning, and realizes the fair and efficient coexistence of LTE-U and Wi-Fi. The simulation results show that the proposed algorithm improves the total throughput of LTE-U and Wi-Fi, and effectively reduces the interference from LTE-U base station to surrounding Wi-Fi users.