Opportunistic Strategy In Mobile Relay Systems

Comparing with the conventional relay system with nearly no mobility, the most significant characteristic of the mobile relay system is the fast and large change of the wireless channel. This characteristic will have both positive and negative effects on the system. For the negative side, the multi-path propagation and serious doppler spreading will induce complex interferences in the system, and degrade the received signal quality. Meanwhile, the shorten channel coherent time will require more frequent channel estimation and feedback hence burden the system. On the other hand, during the communication, we can wait for a better channel condition for data transmission through introducing the buffer-aided strategy. Hence new freedom is introduced into the system which will improve the performance. This is the positive effect of high mobility. Motivated by the positive and negative effects of mobility, this thesis mainly focuses on the fundamental three-node relay system and investigates how to sufficiently exploit the opportunity brought by mobility, i.e., to optimize the system performance while overcoming the negative effect of the mobility.In the first work of the thesis, we study a fundamental problem that how to balance the above two conflicting effects of mobility. Explicitly, we investigate how to make use of the mobility most efficiently to get optimal system throughput under given delay constraint. In par-ticular, a three-node mobile relay system where a relay node moves linearly from the source to the destination is considered. An opportunistic relaying strategy is proposed, which optimally schedules transmission between the source and relay so that the system throughput is maximized while satisfying the delay requirement. Furthermore, the performance of the proposed strategy is analyzed. Especially, the asymptotic throughput achieved by the strategy is derived in the high transmit power regime. Finally, numerical results are given to illustrate the performance of the proposed strategy. Simulation results show its obvious advantages, and well illustrate both the positive and negative effects brought by the relay mobility. This work sheds some light to the design of delay constrained mobile relay system.Note that in the first work, we just consider the change of large-scale fading brought by the mobility. In the second work of the thesis, in order to further exploit the communication oppor-tunity brought by the mobility, we model the wireless channel as a two-dimensional finite state Markov chain (2D-FSMC) model, which can reflect both the large and small scale channel fading in the mobile environment. We still consider a three-node system with one mobile relay, where, however, the relay can have an arbitrary trajectory. For this system, we study how to efficiently schedule the transmission of the source and relay to minimize the energy consumption under a data throughput constraint. The scheduling problem is formulated as a constrained Markov deci-sion process (MDP) and is approximately solved through Lagrange relaxation approach. Based on this, we proposed a probabilistic algorithm, called the opportunistic packet scheduling (OPS) algorithm, to further reduce the energy consumption. The proposed OPS algorithm is proved to be optimal with polynomial complexity. Simulation results show that the proposed algorithm generally outperforms the conventional ones in terms of energy consumption.Finally, for the block fading channel under high mobility, we propose a very practical and optimal relay transmission protocol, which can be adopted generally for an arbitrary mobility model. With this protocol, the system decides the transmitting node based on the channel state information. By scheduling which node to transmit in each slot, the average system throughput is maximized. We further prove that the optimal scheduling decision is only based on the current channel state information and the statistical channel information, which is easy to be realized in the practical system. The proposed protocol can enhance the throughout performance of the system by exploiting the fluctuation of both the large-scale fading and the small-scale fading. Moreover, we propose threshold optimization algorithms for different decision functions as well as a simple method to obtain the optimal thresholds based on the relationship among the elements of the average link SNR sequence. Finally, simulation results show that the proposed protocol outperforms the conventional schemes. Furthermore, it achieves a considerable performance gain even under a finite interval length.