QoS-driven adaptive resource allocation for mobile wireless communications and networks

摘要

Quality-of-service (QoS) guarantees will play a critically important role in futuremobile wireless networks. In this dissertation, we study a set of QoS-driven resourceallocation problems for mobile wireless communications and networks.In the first part of this dissertation, we investigate resource allocation schemesfor statistical QoS provisioning. The schemes aim at maximizing the system/networkthroughput subject to a given queuing delay constraint. To achieve this goal, weintegrate the information theory with the concept of effective capacity and developa unified framework for resource allocation. Applying the above framework, we con-sider a number of system infrastructures, including single channel, parallel channel,cellular, and cooperative relay systems and networks, respectively. In addition, wealso investigate the impact of imperfect channel-state information (CSI) on QoS pro-visioning. The resource allocation problems can be solved e?ciently by the convexoptimization approach, where closed-form allocation policies are obtained for differentapplication scenarios.Our analyses reveal an important fact that there exists a fundamental tradeoffbetween throughput and QoS provisioning. In particular, when the delay constraintbecomes loose, the optimal resource allocation policy converges to the water-fillingscheme, where ergodic capacity can be achieved. On the other hand, when theQoS constraint gets stringent, the optimal policy converges to the channel inversion scheme under which the system operates at a constant rate and the zero-outagecapacity can be achieved.In the second part of this dissertation, we study adaptive antenna selection formultiple-input-multiple-output (MIMO) communication systems. System resourcessuch as subcarriers, antennas and power are allocated dynamically to minimize thesymbol-error rate (SER), which is the key QoS metric at the physical layer. Wepropose a selection diversity scheme for MIMO multicarrier direct-sequence code-division-multiple-access (MC DS-CDMA) systems and analyze the error performanceof the system when considering CSI feedback delay and feedback errors. Moreover,we propose a joint antenna selection and power allocation scheme for space-timeblock code (STBC) systems. The error performance is derived when taking the CSIfeedback delay into account. Our numerical results show that when feedback delaycomes into play, a tradeoff between performance and robustness can be achieved bydynamically allocating power across transmit antennas.