Authors : Ibrahim A. Altawil , Mohammad H. Bataineh and Deifallah A. Dajeh
Abstract: This study discusses the mapping of an important climatic parameter which is crucial in statistical path loss procedures adopted by the International Telecommunications Union (ITU). This parameter is the probability that the refractivity in the lowest 100 m of the atmosphere is less than -100 N-units/km (denoted by � 0). This parameter is available where radiosound station collect their measurements. The interest is to be able to give a value to this indicative parameter` ideally �everywhere in the world. The available `row` data are contour plotted on a world map. Improvements on these maps is suggested by griding the world into �small� squares and evaluating the value at each corner of the square by interpolating with adjacent values. Artificial neural networks is used also as a processing technique to approximate the function representative of the � 0 data so as to develop models by which the values of � 0 could be estimated in locations where data is not available. The technique has been used successfully to estimate � 0 values in the North Western quarter of the hemisphere taken, for example. The results shown in this paper are based on data collected from about 4400 locations throughout the world for the years 1983-1992. The paper compares and contrasts the neural network approach to a suggested estimation methodology and it is shown that good agreement is obtained.
Ibrahim A. Altawil , Mohammad H. Bataineh and Deifallah A. Dajeh , 2006. Radiosonde Refractivity Data Analysis Using Neural Network Techniques. International Journal of Soft Computing, 1: 215-219.