The objective of this project is to advance the field of compressive sensing and broaden its scope into a multitude of new applications, such as ad hoc networking, by significantly enhancing the efficiency through the utilization of problem-specific signals and systems properties. The proposed approach is to design a novel rateless-coding-inspired generalized compressive sensing (GCS) framework and to integrate it into different stages of sensing, information processing, and cross-layer designs in the network design processes.
Intellectual Merit: The proposed research bridges the gap between the advanced theoretical research on modern coding and compressive sensing, as to significantly enhance the efficiency of data acquisition and ad hoc networking. GCS offers the flexibility of exploiting problem-specific properties, such as non-uniform sparsity and non-uniform importance of data in the sensing and recovery phases. This is expected to result in a significant reduction in sensing and communication costs. Through cross-layer design of GCS and underlying network protocols, efficient data acquisition and distributed data storage problems in sensor networks will be investigated.
Broader Impacts: This proposal is expected to have a significant direct impact on advancing the theory and practice of compressive sensing and an indirect impact on many areas in which compressive sensing has potential application, such as medical imaging, hyper-spectral imaging, and bio-informatics. Furthermore, it expected to have an impact on increasing the participation of under-represented students in the PI's research group, as well as in promoting engineering education among high school students, in integration of research and education through curriculum development/revision, in training graduate students, and in promoting multidisciplinary research and collaboration through initiating ECE Seminar Day Program at OSU.
