Discussion

The only source of ambiguity in this technique was introduced by the quantization noise of the received frequency. Sampling the received frequency addressed this problem, but there was a tradeoff between accuracy and usefulness of the derivative. Large sample period generated smoother derivatives but distorted the amplitude information, while short sample periods could not be used to determine the maximum amplitude. For these reasons, a sample of period of 50 was found to be a good compromise. This quantization noise introduced ambiguities in both the location of the peak of the derivative and in the value of the peak of the derivative. These ambiguities correspond to ambiguities in the latitude and longitude of the EPIRB respectively. By taking a range of peak locations and values, our rms (root mean squared) confidence interval was determined to be 150 km or approximately 1.3 degrees. The confidence interval was determined by varying the quantization value and calculating the rms difference for the distances.