As the cellular and PC worlds collide with Wireless LANs and Internet-based packet data, new networking approaches will support the integration of voice and data on the composite infrastructure of cellular base stations and Ethernet-based wireless access points. A cross-layer networking is where the physical and Medium Access Control (MAC) layer knowledge of the wireless medium is shared with higher layers in order to provide efficient methods of allocating network resources and applications over the Internet. With the rapid growth of the Internet and other Internet-protocol related applications, many improvements have been made in the data link layer in wireless systems to support a wider variety of data services. However, further improvements in packet data admission, scheduling and policing between nodes in the network are necessary to maximize throughput and user satisfaction. In the above three, to facilitate diverse QoS support and to provide user satisfaction, scheduling has the greatest impact on system throughput. In the two broad categories of scheduling algorithms like priority and fair scheduling, a majority of them focus on either minimizing packet delay or maximizing user throughput. In this paper, an adaptive cross-layer priority and fair scheduling algorithms have been proposed for multiclass data services that can outperform with respect to user throughput and packet delay. It is also shown that a better performance measure is achieved when validated with NS2 simulations.

Wireless communication is an emerging field and is becoming an essential feature of everyday life. Supporting realtime data flows like audio and video with throughput, delay and fairness constraints would be an important challenge for future wireless networks. The voice service is the only one Quality of Service (QoS) profile dominated the first and second generation wireless systems. Due to the rapid growth of the Internet and Internet-protocol services, the design of recent generation wireless systems have been modified to support a variety of data applications from non-realtime background traffic to streaming video. The above such modifications have resulted in many improvements at the data link layer which facilitates diverse QoS support.

The increasing importance of realtime applications demands provision of QoS and optimal channel access among multiple flows over a shared, error-prone wireless channel. In wired networks, many scheduling methods have been proposed by Pang et al. (1999) and Johnsson and Cox (2001) to bound delays of packet transmission. However, wireless channels are characterized by three features that distinguish themselves from wired networks: (1) serious burst errors, (2) location dependent errors, and (3) multirate communication capability. Burst errors may break data flow's services, while location dependent errors may let error-free flows to receive more services than they deserve. A wireless channel may provide different transmission rates to stations depending on channel qualities. Due to these reasons, Sau and Scholefield (1998) reported that the existing wired solutions may not be suitable for wireless networks.

The cross-layer design methodology has attracted much interest in recent years that leverages the synergy existing between different layers of communication protocol stack to achieve more efficient designs, instead of treating each layer as an individual entity. QoS and channel-aware packet scheduling are important illustrations of the cross-layer design approach which exploit interactions between the physical and the upper layers. A goal of any wireless scheduling scheme is to balance the user's QoS requirements.

Telecommunications Journal, Medium Access Control, MAC, Internet-protocol services, Wireless communication, Wireless Local Area Network, WLAN, Medium Access Control, MAC, Distributed Coordination Function, DCF, Global System for Mobile, GXSM, Inter Frame Space, IFS, Constant Bit Rate, CBR.