A persistent obstacle in conventional signal processing is in achieving well-timed synchronization between analog processing at the sensor front end through to digital signal processing. As improved circuit, sensor and computing technology result in higher carrier frequencies, accuracy of this timing becomes a fundamental limit. Our research seeks to develop a signal processing paradigm that will side-step this limit via detection and estimation with complementary statistics. This processing paradigm will be able to detect and interpret sensor signals that are not captured now, will be well-matched to the analog domain, and will provide for seamless partitioning between analog and digital signal processing.