| dc.description.abstract | Increased utilization of the radio frequency spectrum motivates receivers with wide instantaneous bandwidth for applications such as signals intelligence, cognitive radio, and compliance testing. Traditional wideband receivers inherently have high data rates that are difficult to process and store, and receivers that use multiple analog to digital converters to achieve wide bandwidth have high power usage and cost. Compressive sensing (CS) provides a potential low-data-rate and low-power solution in environments where only a small portion of the wide spectrum monitored is in use at one time, through sub-Nyquist sampling at the information rate. The Nyquist Folding Receiver (NYFR), proposed by Fudge et al., is one such promising CS architecture. The NYFR significantly undersamples input signals causing them to alias such that original frequencies would normally be lost, but the NYFR encodes the original frequency as modulation on the input signals so that all of the original information is preserved. This thesis adds a model of a real-valued NYFR, and uses the model to investigate the design trade-offs inherent to any NYFR receiver. Basic applications including pulse detection, angle of arrival estimation (DoA), and processing of communication signals are simulated. Finally, a prototype receiver was used to experimentally demonstrate the capabilities of a NYFR with an instantaneous bandwidth of 18 GHz while only sampling at 1.5 GSPS. | en_US |
