A low-cost software correlator based on a new polarity correlation algorithm

A new software correlator for polarity correlation estimation is presented. This correlator achieves a high sampling rate by block sampling, which makes it possible to store the signal samples in the individual bits of the microprocessor memory locations, so reducing the overall memory requirements of the correlator. The computational efficiency of the algorithm is derived from the ability of microprocessors to process complete bytes of data simultaneously. This correlator is implemented on a standard 8-bit microprocessor. The algorithm used makes this correlator superior to other software correlators, including the zero-crossing correlator. The latter is, perhaps, the fastest software correlator available. It will be shown that for practical values of sampling rates, the new correlator is more economic in terms of signal storage, faster in terms of time of execution and has an operational bandwidth an order of magnitude higher than that of the zero-crossing correlator.