A novel method for optimal sampled data selection, transmission and reconstruction: a proposal

A technique that reduces the number of samples needed for reconstructing the signal is proposed. The signal is sampled above its nyquist rate and from the sampled data, essential samples, called as key samples, are identified and picked out. Three different implementations using analog circuits, digital circuits and a microprocessor based circuit are reported and their performances are compared. The key samples are provided by the circuit, together with a clock count data and a marking signal. The count data indicates the delay of the present key sample from the occurrence of the previous key sample. A signal reconstructing circuit for the optimal sampler also is reported. The validity of the technique is verified with a computer generated test signal. The proposed technique therefore allows data compression and reduces storage and transmission requirements