Investigation Of Handover And Multiple-Antenna Techniques In Wireless Communications For High Speed Railways

In recent years, the high speed railway (HSR) in China is developing rapidly with the highest operational train speed and the longest mileage in the world. High speed train has brought great convenience to the passengers, and the HSR broadband wireless communication becomes an important and challenging requirements. On one hand, the support of wireless communication techniques is vital for the reliable and efficient train operation, while the passengers increasingly demand for HSR wireless communications service during the trip. On the other hand, there are a number of technical challenges to be tackled in HSR scenarios, i.e., the communication quality is severely affected by the Doppler shift; the received signal strength is seriously degraded due to the penetration loss of carriages; the probability of call drop is increased due to the frequent handovers between cells. In this thesis, based on the "base station (BS)-onboard relay-users" two links architecture, the handover schemes and multiple antenna technologies are studied under the high speed railway scenario.Firstly, the performance of the handover is analyzed, by considering the effect of ground distributed antenna system (DAS), and an optimal handover scheme based on power allocation (PA) is proposed. The handover performance in centralized antenna system (CAS) and distributed antenna system (DAS) with different power allocation is compared, in terms of the probability of handover success, link outage probability and the number of handovers. The results show that, the number of handovers in DAS is significantly less than that in CAS, and the Ping-Pang handover phenomenon is efficiently reduced, meanwhile the probability of handover failure in DAS is much smaller than that in CAS, even without the PA. In the DAS, the geographically difference of each remote access unit (RAU) can provide macro diversity for the system. By utilizing this macro diversity, an optimal scheme with power allocation among RAUs is proposed, which can improve the handover performance by enhancing the received signal strength (RSS) of the onboard relay. The simulation results show that, compared with the equal power allocation (EPA) scheme, the proposed optimal power allocation scheme can achieve a higher probability of handover success and a lower probability of link outages.The linear topology of HSR communication systems leads to the fact that the trans-mission rate has high dependence on the distance between the train and the BS, thus the transmission rate shall change rapidly at different location. Assisted by the train control system (TCS), the BS has the ability of predicting the train position, which makes it possible to apply PA along the railway. The existing researches either only consider the transmission delay, or only consider the system capacity, however, we seek to find a PA scheme to achieve the trade-off between the transmission rate and the delay. Different user services have different requirements on the communication systems, e.g., the voice service requires high real-time capability but less capacity, while data service has a high tolerance on the delay but a high demand on the capacity. In order to satisfy the demands of different type of services and achieve the trade-off between delay and capacity, a PA algorithm is proposed to maximize the ergodic capacity on the condition that the delay performance is ensured. The simulation results show that the proposed PA scheme can significantly reduce the delay at the cost of capacity, which provides some insights on the optimal design of HSR communication system.Then, the selection of cooperation sets is investigated in the HSR communication system based on the DAS on ground and mobile relay with massive antennas. In the train operation, the capacity changes as the distance between the onboard relay and each RAU, and the number of cooperating RAUs has important effect on the average system capacity. The expressions for the instantaneous capacity in a certain position and the average capacity in a cell are derived, and two selection schemes for cooperating RAUs sets are proposed based on the convex optimization theory. One scheme changes the size of RAU sets according to the train position, and achieve the optimal capacity at all positions. However, the switch of RAU sets has high operational complexity due to the high train velocity, the other scheme is proposed with a fixed number of RAUs which is more practical, for which an algorithm is proposed to find the optimal number of cooperating RAUs. It is shown via extensive Monte-Carlo simulations that the proposed algorithm can significantly improve the system capacity.Lastly, we study the optimizing problem on the number of antennas in the onboard relay. In the HSR wireless communication system, the train has a certain length with several carriages, and the antennas of the mobile relay can be centralized on one carriage or distributed on several carriages. The effects of the train speed, the large scale fading, the small scale fading and the power loss of radio frequency (RF) links on the system capacity are studied by adopting centralized and distributed antennas. Meanwhile, the instantaneous capacity at different positions and the average system capacity are derived. Our analysis shows that, compared with the centralized antenna system, the distributed antenna system has a lower capacity near the base station but a higher capacity at the cell edge. It is shown that the mobile relay with distributed antennas can improve the stability of communication services. When the distributed antenna system is adopted in the onboard relay, an algorithm is proposed to optimize the number of antennas in the mobile relay. The analysis and simulation results show that the proposed algorithm can improve the average system capacity.