A unified approach to design of high performance wireless data networks: Power control, transmit optimization, and QOS support

The promise of seamless access to information and multimedia content anywhere and anytime requires the design and deployment of high performance wireless data networks. Some critical features that next generation wireless networking technologies should have include scalability, high bandwidth efficiency, low power consumption and ability to support differentiated services. Scalability of a system requires the algorithms to be distributed, robust, and simple to implement.; To achieve the above goals, this research has developed a general framework in which both cellular systems and Wireless Local Area Networks can be modeled and studied. We analyze a system model consisting of multiple wireless links (each link consisting of a transmitter and a receiver) sharing the same spectrum. The highlights of the new model are that it is packet based, and can handle time varying channel conditions. Contrary to previous studies on the design of distributed algorithms for wireless networks, based mostly on Perron-Frobenius theory, we come up with a novel strategy for design of distributed algorithms. First, we present simple power control formulations which yield optimal solutions having certain structural properties. Based on these structural properties, we design a new family of distributed and asynchronous PCMA (Power Controlled Multiple Access) algorithms. Then, we propose MDMA (Multimodal Dynamic Multiple Access) algorithms wherein the transmitter can dynamically switch between distinct transmission modes in order to match the wireless channel better and deliver packets to the receiver with higher probability of success. We illustrate with a simple example the dynamics of the PCMA and MDMA algorithms and show how they perform better than the constant signal-to-interference-and-noise-ratio algorithm which is used as a benchmark.; We then adapt these new distributed algorithms, which are power and bandwidth efficient, to multi-cellular TDMA/FDMA systems. The simulation studies show power savings of orders of magnitude over the constant SINR algorithm. We also demonstrate how our algorithms can provide differentiated quality of service to users having different delay requirements. We also present novel power balancing algorithms for implementation in a system, where the transmitter has to transmit to multiple receivers, and there is a global constraint on the total power transmitted.