In this project, theory and applications of efficient approximations of the Channel State Information (CSI) in Wi-Fi networks, which describes the key characteristics of the wireless links, will be developed. The approximation will exploit the underlying structure of the CSI and will require very few parameters, resulting in highly effective CSI compression and measurement methods, as well as new data transmission techniques, which will significantly improve the efficiency of Wi-Fi networks and enable more applications and more opportunities in education, public health, and business that increasingly depend on the high speed and reliability of wireless networks. Fundamental issues that will be addressed include: finding the theoretical explanation of the approximation, designing fast CSI compression algorithms, designing more efficient CSI measurement and prediction methods, and designing novel data transmission techniques. The knowledge gained in this project will advance the research field by revealing important features of a large class of wireless channels that were previously unnoticed, and developing optimized methods for such channels. Results obtained in this project will be used in classes related to networking. Both graduate and undergraduate students will participate in this project, and students from underrepresented and minority groups will be actively reached out to and recruited.

This project is motivated by an interesting experimental discovery, which shows that the CSI vectors in Wi-Fi networks can be approximated very well in many case as the linear combination of very few, such as 3, sinusoids, even when the number non-negligible paths are much larger. Referring to channels with such SParse Sinusoid (SPS) approximation as SPS channels, the goals of this project include: 1) understanding the theoretical foundation regarding to the existence of the SPS approximation by studying the mathematical properties of channel and designing fast CSI compression algorithms, 2) designing efficient CSI measurement and prediction methods by exploiting underlying structure of the SPS approximation, and 3) designing new data modulation techniques for SPS channels by exploiting the simplified representation of the channel. The proposed algorithms and techniques will be implemented in experimental platforms and tested in real-world wireless channels. By removing the bottleneck caused by the high overhead in CSI feedback and measurement, the outcomes of this research will be timely solutions for Wi-Fi networks for better supporting MU-MIMO or massive MIMO. The new data modulation techniques for SPS channels will likely improve the link speed while reducing the complexity.