Muiti-layer optimization in wireless ad hoc networks

Abstract

The growing prevalence of wireless ad hoc networks calls for an innovative design to support Quality of Service (QoS) while maintaining high energy efficiency and bandwidth efficiency. In this dissertation, a multi-layer optimization approach is developed in view of benefits and necessities of sharing information among different layers. Specifically, given the traffic demands with QoS requirements, how to jointly design physical, MAC and network layers to optimize the network performance is considered in this dissertation. Firstly, a joint power control and maximally disjoint multipath routing scheme is proposed for QoS provisioning of end-to-end traffic with minimum rate constraint. A framework of power control with QoS constraints is introduced and both centralized and distributed solutions are derived. It is demonstrated by simulations that the proposed scheme provides high energy efficiency and the prolonged network lifetime, as well as robustness when augmented with a dynamic traffic monitoring and switching mechanism. In order to fulfill the QoS requirement at the link layer, TD/CDMA has been chosen as the MAC scheme due to its support for a high network throughput in a multihop environment. The multi-link versions of proportional fair and throughput optimal scheduling algorithms are proposed for multihop wireless ad hoc networks. In addition, a generic token counter mechanism is employed to satisfy the minimum and maximum rate requirements. Approximative algorithms are suggested to reduce the computational complexity. In networks that lack centralized control, distributed scheduling algorithms are derived and fully distributed implementations are provided. Simulation results demonstrate the effectiveness of the proposed schemes. In order to further improve bandwidth efficiency, cognitive radio is considered for more efficient spatial and temporal spectrum sharing. Specifically, we consider the scenario where a cognitive radio ad hoc network is formed by low power personal/portable devices operating simultaneously in the same frequency band along with a legacy system. A power control problem is formulated to maximize the energy efficiency of the ad hoc network, as well as to guarantee QoS for both legacy network users and ad hoc network users. The results show that cognitive radio greatly improves bandwidth efficiency of wireless ad hoc networks.%%%%