Proactive Adaptive Channel Reconfiguration (PACR)

Wireless networks, particularly ad hoc networks, are vulnerable to adverse impacts from dynamic network conditions, RF environments, network congestion, limited power supplies, etc. As critical applications move to wireless networks, it is necessary to predict and avoid these adverse impacts before they occur. Methods of network optimization based on research in cross-layer optimization and cognitive networks have the potential to improve the reliability of wireless networks. We will investigate cross-layer optimization techniques which allow the network to proactively respond to deteriorating or improving conditions at any layer before the conditions affect network applications. A framework for deploying and implementing algorithms for proactive optimization at the application-layer will be modeled. As an option, a limited version of the framework will be demonstrated in a live 802.11 network.