NSF Award
1520496
The broader impact/commercial potential of this project will be to develop a scalable framework for increasing the spectral efficiency of wireless Multiple-In, Multiple-Out (MIMO) systems in environments with long channel coherence time by leveraging implicit channel sounding. To date, standards committees and equipment vendors have avoided implicit channel sounding since key technical questions regarding its use remained unanswered. By providing a zero-overhead method for implicit channel sounding, the proposed system will be able to provide rich environment and user mobility data to the Media Access Controller (MAC) and application layer, with strong implications for the future design of rate and group selection for MU-MIMO systems. With commercialization targeted at unlicensed TVWS wireless devices, the proposed 4-antenna system will be able to provide at least a 4-fold increase in network capacity compared to existing single-antenna TVWS solutions, without suffering from protocol overhead congestion typical in existing multi-user MIMO systems. Integral to this approach is the ability to implement the technology on low-cost software-defined radio hardware, placing this technology within the reach of underserved communities looking for low-cost broadband data solutions.<br/><br/>This Small Business Innovation Research (SBIR) Phase I project will demonstrate an innovative implicit beamforming protocol that eliminates channel sounding overhead in TV-band White Space (TVWS) radio channels, enabling efficient and high-speed, last-mile unlicensed wireless network deployments. The project will develop and benchmark practical implicit channel reciprocity calibration for software-defined radio arrays, while utilizing an innovative software-defined radio signal flow that allows the system to be implemented on low-cost commodity system-on-chip hardware. Both approaches represent a departure from existing methods of beamforming in commercial systems, which rely on both overhead-intensive explicit beamforming and real-time parallel hardware for signal processing and beamforming calculation. While shown to be feasible in academic research papers, both key innovations have yet to be validated in an operational system and several known barriers to implementation are investigated in this project. Finally, the proposed implicit beamforming system will be validated on real TVWS channels though trace-driven system emulation. To our knowledge, this is the first time that an implicit beamforming system will be tested on commercial hardware and on real wireless channels.
