Efficient Signal Recovery from Level Crossing Sampling Method

Abstract A novel level-crossing (LC) algorithm, termed reduced LC (RLC) sampling, is presented that mitigates the effect of oversampling in LC analogue-to-digital converters. Unlike algorithms that modify bit-depth to control oversampling, RLC uses a simple rate modification method that can be implemented using minor additional circuitry. Additionally, a new signal reconstruction method is developed that guarantees stable reconstruction from a sparse sample set. Jointly, these algorithms enable alias-free signal reconstruction when sample rates are well below the Nyquist limit.