Columbia University

NSF-BSF grant to Improve Wireless Networks Robustness via Weather-Sensitive Predictive Management

image001Prof. Zussman, Prof. Hagit Messer Yaron (Tel Aviv University), and WiMNet Postdoctoral Research Scientist Dr. Jonathan Ostrometzky received an NSF-BSF grant titled “Improving Wireless Networks Robustness via Weather-Sensitive Predictive Management”.

The $580K project which is co-funded by the NSF and the U.S.-Israel Binational Science Foundation (BSF) focuses on the wireless networks that are used in the backbone of cellular, smart cities, and emerging 5G networks. These networks rely on millimeter-wave (mmWave) frequencies, which are sensitive to weather conditions and specifically to rain events. Improving resilience to such events, the rapid increase in wireless traffic, and the Quality of Service (QoS) demands of mission-critical smart city applications, all call for dynamic network management schemes. Therefore, weather-sensitive network control and management approaches will be developed, aiming to improve network resilience and performance. The innovation of this project is to use weather-affected measurements of wireless link states in the network to accurately predict their future states and to provide input to network control schemes. These schemes include adjustments of links’ modes and network topology to the moving rain, prior to its effects on the signals. The algorithms designed will build on extensive datasets of wireless links and the algorithms will be evaluated and demonstrated in the NSF PAWR COSMOS testbed.

Specifically, the project focuses on backhaul and fronthaul networks which are currently transitioning to E-band (60-90 gigahertz) links that are very sensitive to rain events. Contrary to legacy (4G) cellular networks where local physical layer adaptation has been sufficient, in emerging smart city and 5G networks (that will require low latency and high bandwidth), link and network layer adaptations will be essential. Algorithms that use the self-extracted attenuation measurements from the network to predict the channel states throughout the network will be developed based on relationships between weather and signal attenuation. Then, weather-sensitive cross-layered control algorithms will be developed. These algorithms will jointly optimize power, modulation and coding, channel allocation, and routing to satisfy QoS requirements in response to predicted changes in network conditions. Finally, the project’s contributions will include analysis of first-of-their-kind mmWave backhaul measurements from a smart city network in Israel and unique evaluation in a city-scale testbed that integrates first-of-their-kind mmWave transceivers.