Interference between neighboring wireless networks is a serious problem in high-density deployments. As one well-known example, WiFi networks in dense deployment areas can experience severe performance degradation. Current approaches efficiently share available bandwidth between networks, but do not address the underlying interference problem. One potential approach is multiple-input multiple-output (MIMO) interference cancellation. While the theory of MIMO interference cancellation is well understood, it has not been deployed due to practical limitations. This project is designing, evaluating, and implementing practical solutions to MIMO interference cancellation. Ideas being studied include Consistent Access Point Orientation to reduce the complexity of gathering channel state information, Proportional Fairness with Interference, which enables better MIMO link scheduling, multi-level scheduling to reduce computational overheads, tit-for-tat policies that encourage access point cooperation, and Guided CSMA/CA to guide nodes toward a scheduled execution. Approaches are being implemented and evaluated within the ns-3 network simulator using accurate physical layer models, and are also being tested on a software-defined radio platform.
This project aims to significantly boost performance of wireless networks, such as WiFi networks and femtocells, that operate in dense deployment areas. The approach taken enables each network to use the full available bandwidth of the wireless channel regardless of surrounding interference. The project is expected to produce algorithms and protocols to enable MIMO interference cancellation techniques to be adopted in practical network settings. The project will also contribute to the research infrastructure by implementing support for MIMO interference cancellation the ns-3 network simulator and making the enhancements available to the research community.
