With its large-amplitude and short-duration pulses, atmospheric noise, produced by electrical discharges within clouds, dominates the low-frequency spectrum. Unless mitigated, this noise, which enters into the Loran band, can distort a Loran signal within a receiver and can induce cycle selection errors resulting in range measurement errors of 3,000 km. Such errors would induce position errors greater than 556 m and would prevent the receiver from meeting the requirements for aircraft non-precision approach (NPA). In order to evaluate the effectiveness of Loran for non-precision approach we needed to perform two tasks: 1) confirm that the standard model of atmospheric noise, ITU P372-7, despite its caveats against its use in the low-frequency band is indeed valid for Loran, and 2) obtain raw data for use in evaluating non-linear signal processing techniques to mitigate the effects of atmospheric noise. To accomplish these tasks, we developed atmospheric noise collection equipment and fielded them in Norman, Oklahoma, a location of high storm activity. Since atmospheric noise can be several orders of magnitude larger than a weak Loran signal, our latest receiver design combines a high gain and low gain channel to provide 122 dB of dynamic range while having 12 bits of resolution for the Loran signal. This paper describes our work on verifying the accuracy of the ITU atmospheric noise model for both long-term and short-term noise. In addition, we extend the ITU model by showing a correlation between the predicted rms noise envelope field strength and the minimum voltage deviation of the noise. These results are important to determining the processing credit for non-linear signal processing. The processing credit can then be use to more accurately show the coverage and availability of Loran for aircraft non-precision approach (NPA).
展开▼
