Abstract: Techniques are provided for compressive sensing (CS) in a sensor network, for improved power efficiency. A methodology implementing the techniques according to an embodiment includes determining a state of a sensor network, based on a calculated statistic of sampled data values generated by one or more sensors in the network, and on anomaly indications generated by the one or more sensors. The method further includes calculating a CS sampling schedule based on the determined state and further based on a sparse signal recovery algorithm. The method further includes broadcasting the CS schedule to the one or more sensors. The CS schedule includes a sensor identification, sampling frequency, and sampling time offset for each sensor to be sampled. The method further includes updating the state of the sensor network and the CS schedule, based on updated data values generated by the one or more sensors in accordance with the sampling schedule.

Abstract: Techniques are provided for compressive sensing (CS) in a sensor network, for improved power efficiency. A methodology implementing the techniques according to an embodiment includes determining a state of a sensor network, based on a calculated statistic of sampled data values generated by one or more sensors in the network, and on anomaly indications generated by the one or more sensors. The method further includes calculating a CS sampling schedule based on the determined state and further based on a sparse signal recovery algorithm. The method further includes broadcasting the CS schedule to the one or more sensors. The CS schedule includes a sensor identification, sampling frequency, and sampling time offset for each sensor to be sampled. The method further includes updating the state of the sensor network and the CS schedule, based on updated data values generated by the one or more sensors in accordance with the sampling schedule.