Date of Award
Doctor of Philosophy
Electrical and Computer Engineering
Compressive sensing (CS) technique potentially allows sparse signals to be sampled at rates lower than their Nyquist Rates, making it appealing for implementation of low-power sensors. This dissertation investigates techniques to further improve CS efficiency by adaptively adjusting the sampling rates or CS measurement sizes according to signal sparsity variations. The proposed technique is referred to as adaptive compressive sensing (ACS). The work first investigates the feasibility to perform ACS operation. The study indicates that the sparsity of many biomedical signals does change over the time and therefore such adaptive operations are feasible. This study further investigates the potential power saving by the proposed ACS operations and it shows significant power reduction, up to 11.02 to 32.18 percent, can be potentially achieved by the proposed ACS operation, in wireless sensor nodes whose power consumption is dominated by the transmitter power consumption. Finally, the potentials of using analog wavelet filter circuits to monitor the changes of signal sparsity is investigated. The capability to detect signal sparsity changes at the sensor node is critical for the practical implementation of the proposed ACS operation. The proposed sparsity detection circuit comprise of low-power analog wavelet circuits and additional simple circuits that count the number of threshold crossing at the output of the analog wavelet circuits. Simulation results demonstrate promising potentials of the proposed method.
This dissertation is only
available for download to the SIUC community. Others should contact the
interlibrary loan department of your local library or contact ProQuest's Dissertation Express service.