Study On Access Technology Of Unlicensed Bands

LTE (Long Term Evolution) is a widely used mobile communication system all over the world. In order to satisfy the increasing mobile data traffic, LTE system targets at coexistence with other systems in unlicensed bands. To guarantee the coexistence between LTE and other wireless technologies in unlicensed bands like Wi-Fi, 3GPP (The 3rd Generation Partnership Project) proposes LAA-LTE (Licensed-Assisted Access Using LTE) as the key technology of access to unlicensed bands, in which the core scheme is LBT (Listen Before Talk). LBT makes it possible for LTE to sense the channel before occupy unlicensed bands, which guarantees the fairness of the competition to occupy the channel.Now there are some problems in LAA-LTE system, for example:though kinds of practical schemes and corresponding resource allocation strategies exists, a framework of resource allocation is leaked; the framework needs a specific LBT scheme, which can adjust parameters according to the given resource allocation strategy; recent LBT don't support frequency reuse, and user feedback hasn't been considered in improved methods. In connection with the above problems, in this paper we focus on the resource allocation strategy based on traffic balancing,LBT design with dynamic CW (Contention Window) and frequency reuse technology depending on user feedback, and solve the corresponding problems effectively.To achieve a resource allocation framework in unlicensed bands,problem is formulated by considering the power allocation and occupancy ratio allocation at the same time based on traffic balancing, and extends the traditional single base station scenario to multi-base station scenario.Heuristic algorithm is used to solve two allocations respectively, and locally optimal resource allocation framework is achieved. Simulation results show the influence of different parameters to strategy performance,and analyses the characteristics of occupancy ratio under multi-base station scenario.Focus on LBT design which is based on resource allocation framework, in this paper the fundamental of Wi-Fi and LBT is analyzed.Problem formulation is solved by two-dimension Markov chain and the relationship between the length of contention window and occupancy ratio in unlicensed bands is derivate to implement traffic balancing. Compared with other schemes, the scheme improves the user satisfaction.To deal with the LAA frequency reuse problem, a frequency reuse method based on user feedback is proposed. Specifically, the features of current frequency reuse methods are studied, and the performance of frequency reuse is improved by user feedback. A transmission timing alignment strategy is proposed to enhance user throughput, for the contention window is adjusted by NACK signaling of user. Simulation results show that our strategy has 13% gain in user perceived throughput with the contrast algorithm.The whole research is summarized in the last part of the thesis, and further research directions are discussed.