Research On Cross-layer Design With Imperfect CSI For Centralized/Cooperative MIMO Systems

Multiple input multiple output (MIMO) is the use of multiple antennas at both the transmitterand receiver, to improve the capacity and spectrum efficiency of communication system withoutadditional bandwidth or increased transmit power. Cross-layer design makes full use of the correlationof each layer in the network, and realizes adaptive optimal allocation and reasonable resourcesaccording to the dynamic change of wireless network environment. Therefore, the cross-layer designstudy for the MIMO system makes full use of limited wireless resources while improving the systemperformance, which will have important theoretical significance and value of research.In this thesis, for the imperfect channel state information (CSI) caused by the feedback delay,through combing adaptive modulation in the physical layer and truncated automatic repeat request inthe data link layer, the cross-layer optimization in the space-time coded MIMO system and thecooperative MIMO system of relay selection based on the imperfect CSI are investigated in order toimprove the SE of system better, which will provide a theoretical basis for further practical cross-layerdesign. The main contents of the thesis are summarized as follows:1. The cross-layer design based on imperfect CSI in the space-time coded MIMO system isinvestigated, and the average packet error rate (PER), average packet loss rate (PLR) and averagespectral efficiency (ASE) and other expressions of the system are derived. Taking into account theadaptive thresholds affected by the feedback delay, the variable thresholds for AM are proposed basedon the maximum a posteriori probability, the simulation results show that the presented variablethresholds method can effectively reduce the impact of delay on the system performance whencompared to the conventional fixed thresholds method. Thus, the ASE of the system is increased, andthe average PLR is decreased.2. The cross-layer optimization based on imperfect CSI is studied in the MIMO systemcombining the power allocation and space-time coding. In this study, by maximizing the ASE of thesystem, a sub-optimal power allocation based on maximizing the compressed signal to noise ratio isproposed, and the closed-form solution is available, which simplifies the computational complexity ofthe optimal one. The simulation results show that the proposed algorithm has almost the sameperformance as the optimal power allocation algorithm, which greatly improves the ASE of thesystem, and achieves a tradeoff between complexity and performance effectively.3. The cross-layer design in the cooperative MIMO communication system is studied, and the cross-layer optimization scheme is proposed based on imperfect CSI and relay selection, the averagespectral efficiency and packet error rate of the system are respectively derived in order to evaluate theperformance in theory. Based on these, the cross-layer optimization subject to the average PERconstraints is proposed to overcome the performance loss caused by the conventional instantaneousPER constraints. The design optimizes the adaptive thresholds by maximizing the ASE of the system,and proposes an iteration method joint the Lagrange multiplier and Newton methods to obtain thethresholds. The simulation results show that, the proposed optimization scheme can obtain higherspectral efficiency, when compared to the cross-layer design with the instantaneous PER constraints.