Research On Radio Resource Management For Broadband Wireless Communications Under High Mobility Environments

In recent years, high-speed vehicles represented by high-speed railway got rapid development around the world. In order to satisfy the informationization of transportation industry and ubiquitous communica-tion demand, there is a strong demand to research the broadband wireless communications that can be applied to high mobility environments. Most of existing researches about high mobility broadband communica-tion has focused on the broadband access solutions, channel modeling and estimation, correction of Doppler shift, and handoff scheme. The research of radio resource management (RRM) algorithms besides handoff scheme is still in its infancy. However, the RRM is of great importance to improve the system performance. Then, it is significant to conduct research on RRM and design some high-mobility RRM algorithms that suitable for high mobility environments.In this dissertation, we will study the typical example of RRM for broadband communication under high mobility environments, i.e., the RRM for high-speed railway communication (HSRC). Find out the key techniques, and design some RRM algorithms to overcome the problem that traditional RRM algorithms cannot suitable for HSRC systems. Then we get some viewpoints about the methodology of design RRM algorithms for such professional systems, and offer some valuable references to future research of RRM algorithms that suitable for other high mobility environments.In HSRC systems, there are some new challenges for the design of RRM algorithms, including the link adaption algorithm for train-ground link needs to consider the impacts of imperfect channel state information (CSI) and meet the requirement of low computational complexity given by HSRC fast fading channel; the admission control scheme design for train-ground link needs to handle the frequent handoff services with batch arrival and the impact of one user always with a lot of different services; the resource allocation policy for train-carriage communication should consider the impacts of inconenient energy supply, short distance communication, and device-to-device (D2D) communication demand.In this dissertation, for the challenges of link adaptation transmission, admission control and resource allocation in HSRC systems, corresponding approaches are proposed by combining the characteristics of high-mobility scenarios and existing research results. Relevant performance metrics are evaluated by simulation. The main contributions of the dissertation are summarized as follows:1. Base on the partially observable Markov decision process (POMDP), a novel link adaption algorithm is developd for train-ground link to improve system goodput. Firstly, the challenges of desiging the link adaption are investigated, which include the constraint of imperfect CSI and low computational complexity. Secondly, two important transmission parameters i.e., diversity/multiplexing gain in physical layer and frame size in link layer, are introduced; and the expression of goodput is derived. Thirdly, the proposed link adaption algorithm is designed based on POMDP model. Since the characteristics of train-ground link have been fully considered, then the proposed algorithm can choose the optimal transmission parameters with low computional complexity to improve the goodput according to the imperfect CSI and the historical decision informations. Finally, the effectiveness of the proposed link adaptation algorithm is demonstrated by simulation.2. A novel admission control scheme is proposed to HSRC train-ground link, where resource reservation algorithm and resource preemption model are developed together to reduce the service dropping probability. Firstly, the main reasons of on-going service dropping related to admission control scheme are analyzed, which include frequent adjustment of services’modulation-and-coding-scheme (MCS) and continual handoff events with batch arrival. Secondly, using the characteristics of HSRC, the resource requirements of MCS adjustment and handoff service are analysed, then an adaptive resource reservation algorithm is proposed to optimize the amount of reserved bandwidth for on-going services. Thirdly, considering the scenarios of resource insufficient, a spring-based resource preemption model is further developed to reasonably obtain the available bandwidth resources from vary-bit-rate service under the condition of minimizing the the effect of resource preemption on on-going services. Then, according to the above works, the admission control scheme is proposed to optimize the service dropping probability performance. Finally, simulation results are presented to demonstrate the effectiveness of the proposed scheme.3. Considering the characteristics of train-carriage communication, i.e., inconenient energy supply, short distance communication, and D2D communication demand, a novel energy-efficient resource allocation scheme is proposed, where the energy consumptions come from transimitter and receriver are all taken into account. The time-frequency resource unit (RU) and power resources are allocated together to optimize the energy-efficiency performance. Firstly, the energy-efficient resource allocation problem is formulated as a mixed integer and non-convex optimization problem. Then, to reduce the computation cost, we tackle this problem with two steps. Step 1, the RU allocation policy is obtained via a greedy search method, and the original optimization problem is reduced to a non-convex fractional programming problem. Step 2, after obtaining the RU allocation, exploiting some manipulations, we transform the reduced problem to a concave optimization problem, and obtain the sub-optimal power allocation strategy through the Lagrange dual approach. Finally, simulation results are presented to validate the effectiveness of the proposed resource allocation scheme.