Energy-Efficient Chip-Scale Sensing of CrowdedWide Spectrum

点击翻译按钮获取中文摘要

The ever-increasing demand for wireless applications has resulted in an unprecedented radio frequency (RF) spectrum shortage, yet access to spectrum will be an increasingly important foundation for our nation's economic growth and technological leadership. A dynamically shared spectrum access scheme is expected to significantly boost the spectrum efficiency. The key enabling technology for the shared spectrum access is a real-time sensor that can monitor a very wide and crowded spectrum. Unfortunately, real-time access to a wide and crowded spectrum using existing technologies requires extremely power-hungry high-speed analog-to-digital converters and hence is not practical for energy-constrained mobile applications. This project will develop a new generation of energy-efficient and low-cost spectrum sensing systems by fusing recent innovations in RF acoustic-resonator-based devices, wireless circuits, and sparse signal processing. If successful, this new chip-scale affordable system will enable a transformative functionality -- energy-efficient sensing of densely occupied wide spectrum in real time -- that allows us to substantially enhance spectrum efficiency via dynamic spectrum access. In addition, activities are planned to ensure that this project will have long-lasting effects in a number of important dimensions, from education to industry, and from electromagnetic spectrum policy to outreach.Recent work has demonstrated the possibility of acquiring GHz-wide signals with moderate-speed analog-to-digital converters at a sub-Nyquist rate by leveraging compressive sensing or sparse Fourier transform. However, all these systems require the spectrum to be sparse and fall short when the spectrum is densely occupied. In contrast to existing spectrum sensing approaches that digitize the entire bandwidth of each channel, this project harnesses a new idea that monitoring only a small fraction of the channel bandwidth is sufficient for occupancy detection. Based on this concept, the project will leverage a novel overtone micro-electro mechanical-system (MEMS) RF resonator to create many very narrow, sharp, and equally spaced passbands across the GHz-wide spectrum. The end result is that the overtone filter sparsifies the spectrum by suppressing, within every channel, the redundant spectrum content for occupancy detection. This filtering-created sparsity allows the utilization of sparse Fourier transform algorithm and moderate-speed analog-to-digital converters to achieve spectrum sensing with high energy efficiency. With a sufficient number of filter passbands, the proposed energy-efficient spectrum sensor captures all channels and does not miss any occupancy information. This project will also develop a wide-tuning-range RF overtone filtering microsystem through a filter-circuit co-design to support multi-band operations and sensing of multi-GHz wide spectrum via scanning.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.