Cross-Layer-Model Based Adaptive Resource Allocation for Statistical QoS Guarantees in Mobile Wireless Networks

We propose a cross-layer-model based adaptive resource-allocation scheme for the diverse quality-of-service (QoS) guarantees over downlink mobile wireless networks. Our proposed scheme dynamically assigns power-levels and time- slots for heterogeneous real-time mobile users to satisfy the variation of statistical delay-bound QoS requirements. To achieve this goal, we apply Wu and Negi's effective capacity approach to derive the admission-control and power/time-slot allocation algorithms, guaranteeing the statistical delay-bound for heterogeneous mobile users. When designing such an algorithm, we study the impact of physical-layer issues such as adaptive power-control and channel-state information (CSI) feedback delay on the QoS provisioning performance. Through numerical and simulation results, we observe that the adaptive power adaptation has a significant impact on statistical QoS-guarantees. In addition, the analyses indicate that our proposed resource-allocation algorithms are shown to be able to efficiently support the diverse QoS requirements for various real-time mobile users over different wireless channels. Also, in an in-door mobile environment, e.g., the widely used wireless local-area networks (WLAN), our proposed algorithm is shown to be robust to the CSI feedback delay.