Traditional communication systems are designed using a layered approach based on the open system interconnection (OSI) reference model. According to this design philosophy, each layer offers certain services to the higher layers, by shielding those layers from the details of how the services are implemented. Consequently, each layer optimizes its own goal and the design can hardly be optimal from an overall system point of view. Because of the increasing demand of wireless communication systems, it is more and more necessary for the system designers to implement more efficient protocols through a cross-layer approach. The nature of a cross-layer design is to provide an innovative insight into the vertical integration of different protocol layers with the ultimate goal of achieving effic...[ Read more ]

Traditional communication systems are designed using a layered approach based on the open system interconnection (OSI) reference model. According to this design philosophy, each layer offers certain services to the higher layers, by shielding those layers from the details of how the services are implemented. Consequently, each layer optimizes its own goal and the design can hardly be optimal from an overall system point of view. Because of the increasing demand of wireless communication systems, it is more and more necessary for the system designers to implement more efficient protocols through a cross-layer approach. The nature of a cross-layer design is to provide an innovative insight into the vertical integration of different protocol layers with the ultimate goal of achieving efficient management of system resources.

In this thesis, we present several cross-layer methodologies, primarily in the context of integrating the physical (PHY) and medium access control (MAC) layers, to more efficiently support adaptability and optimization in wireless networks. Our numerical and simulation results demonstrate that significant improvement in the system performance such as throughput, power-efficiency, average packet delay, and system stability can be achieved by our cross-layer approaches, compared with conventional schemes where integrated layer adaptation, design, and optimization are not used.

Two types of wireless architectures are investigated: Wireless ad hoc networks and infrastructured wireless networks. For wireless ad hoc networks, we propose a cross-layer optimization framework to jointly design the scheduling and power control. Specifically, the transmitted power and constellation size are dynamically adapted based on the packet arrival, quality of service (QoS) requirements, power limits, and channel conditions. A key feature of the proposed method is that it facilitates a distributed implementation, which is desirable in wireless ad hoc networks. The performance of our proposed methodology is investigated for both cases of unicasting and multicasting.

As for infrastructured wireless networks, we first propose a cross-layer adaptive resource allocation and scheduling approach for the downlink transmission of a multi-input multi-output (MIMO)/orthogonal frequency division multiplexing (OFDM) system. The proposed algorithm jointly implements the scheduling at the MAC layer and the subcarrier, bit and power allocation at the physical layer. We then investigate the uplink counterpart, which jointly designs a multi-packet reception (MPR)-based MAC protocol and adaptive resource allocation for a MIMO/OFDM system. Our algorithm provides interconnection and information exchange between the MAC and PHY layers through the use of a request-to-send (RTS)/clear-to-send (CTS) based multiple access mechanism, and an allocation problem is formulated to encapsulate both of the MAC and PHY issues. Finally, we extend our design by adopting a channel state information (CSI)-based random access in a wireless local area network (WLAN) system supporting MPR. In this algorithm, when the backoff time counter reduces to zero, each node selectively transmits according to a channel threshold which is determined by the network population, estimated CSI, as well as the MPR capability of the system. A throughput expression is also derived by taking transmission errors into consideration. The optimal channel threshold is also obtained to maximize the system throughput.