| Wireless mobile communication has made huge progress in recent years, and multi-hop relay and orthogonal frequency division multiple access (OFDMA) are two important new technologies in such field. Multi-hop relay can achieve higher link capacity and wider wireless network coverage, by dividing one wireless link with poor quality into multiple wireless links with better quality. On the other hand, OFDMA, which is based on orthogonal frequency division multiplexing (OFDM), can not only eliminate inter symbol interference (ISI) by adding appropriate cyclic prefix (CP), but also improve system throughput through multi-user diversity, which adaptively allocates subcarrier and power for different users. Therefore, it is necessary to investigate the key networking issues for next generation wireless multi-hop relay networks, which mainly applies those two new wireless communication technologies above. By using optimization theory, game theory and pricing policy, research issues of next generation wireless multi-hop relay networks such as routing algorithms, handover schemes and radio resource allocation schemes are analyzed and investigated thoroughly in this doctoral dissertation, in order to improve system performance as a whole.Firstly, to increase end-to-end throughput as well as decrease end-to-end delay for wireless multi-hop networks, an AODV routing algorithm which is based on network coding is proposed. Routing algorithm is a key factor for wireless multi-hop networks, while network coding is widely applied as a new communication technology. The proposed routing algorithm make the integration by adding network coding module which includes data queue and timer to original AODV routing algorithm, and also by adding network coding header to the header of data package. The proposed routing algorithm is easy to implement and with little overhead. Simulation results show that, in the setting topology, the proposed routing algorithm can lead to higher network throughput and lower average end-to-end data package delay, hence the network performance is improved. The proposed routing algorithm is designed for the networking layer of common wireless multi-hop networks, and is independent of the actual realization of physical layer and MAC layer in a node, so it can also be applied to next generation wireless multi-hop relay networks which use various technoligies in physical layer as well as in MAC layer.Secondly, a novel handover scheme based on multi-hop relay for GSM-R networks is proposed, focusing on the handover issue which plays an important role in keeping GSM-R networks reliable. Due to Doppler shift of received signals caused by high-speed moving of trains, traditional handover scheme for GSM-R networks fails to achieve high successful handover rate. In the proposed handover scheme, multiple relay stations are set in the overlapping area between adjacent cells to transmit signals which assist mobile stations to handover. Meanwhile, the multiframe in control channel of GSM-R networks is modified accordingly to avoid co-channel interference between signals received from relay stations and from base stations. When a mobile station is in handover process, it not only measures the signal received from neighboring base station, but also has to measure the signal received from the relay station which is set in the overlapping area, before making the final handover decision. It is observed in numerical simulation that, handover outage rate using the proposed handover scheme outperforms that using traditional handover scheme for GSM-R networks, thus increasing the successful handover rate in GSM-R networks.Thirdly, by using optimization theory, resource allocation problem for OFDMA-based wireless multi-hop cellular networks is investigated. Previous research on resource allocation for OFDMA-based wireless networks normally only considers scenario where there is only one cell with single-hop downlink or uplink between base station and mobile station, so those proposed resource allocation algorithms cannot be directly applied to wireless multi-hop cellular networks. In order to reduce the difficulty in discussing resource allocation issue for wireless multi-hop cellular networks, a hierarchical tree topology to describe wireless multi-hop cellular networks is introduced in advance, and such topology is the basis for the following optimization functions. Then, according to the two types of optimization which are rate adaptive (RA) and margin adaptive (MA), two optimization functions based on system throughput and total system transmission power are formulated respectively, as well as constraints for each optimization function are imposed. To reduce calculation complexity, two sub-optimal resource allocation schemes both containing subcarrier allocation and power allocation are proposed respectively in accordance to those two previous optimization functions. Through simulation, it is shown that compared to the fix resource allocation algorithm such as OFDM-TDMA, the proposed resource allocation schemes can improve system throughput or reduce total system transmission power, while keeping proportional fairness among wireless links in the system.Lastly, by using game theory and pricing policy, Resource allocation issue in uplink OFDMA multi-hop relay networks is investigated. With constraints such as maximum transmission power being imposed, a multi-cell optimization model for resource allocation in uplink OFDMA multi-hop relay networks is first formulated. Then, by applying non-cooperative game and pricing theories, the optimization is transformed into the case where power is independently allocated on each sub-channel. The existence and uniqueness of Nash Equilibrium for power allocation model which is based on Non-Cooperative Game is proven, and accordingly a distributed algorithm is proposed. Simulation results show that, the proposed algorithm can improve the system throughput effectively while profoundly reducing total system transmission power, thus achieving much higher energy efficiency ratio.
|
PDF Full Text Request