Abstract: A received signal is sampled and a mean is calculated. A zero crossing sample is then located and used to locate a starting sample and the samples are read from that point forward, using sampling intervals and mean crossings, to decode the bit stream. Alternatively, first and second derivatives of the signal are calculated. A sign (.+-.) of the first derivative sample at the first zero crossing of the second derivative is used to determine the sign of a sample. The second derivative of the signal is then read, using sampling intervals and the sign of the first derivative, to decode the bit stream. In another alternative, samples are used to calculate a rolling average of the samples. The rolling average of the sample is compared to the sample to determine the logic value of the sample based upon whether the sample is above or below the rolling average.