Prediction of dust storms-induced cross polarization at microwave and millimetre wave bands

Abstract

Electromagnetic wave propagation in arid and semi-arid regions is influenced by sand and dust storms (SDS). This has received considerable interest in recent time with emphasis on signal attenuation and cross polarization (XP) in dual polarized systems for optimal conservation of the frequency spectrum. In the face of paucity of well-developed SDS data and unavailability of directly measured propagation parameters (especially cross polarization) and statistics for telecommunication paths, feasible network planning and efficient system design in such regions require prediction of percentage time statistics of propagation parameters from scanty available data. SDS meteorological data obtained from Nigeria is thus statistically processed in an attempt to bridge such gap. Aside from non-sphericity and eccentricity of falling dust particles which are responsible for differential attenuation and phase rotation, XP has been attributed to tendency of the particles to align in a particular direction (canting angle). While the dust particle shape and the phase rotation are yet to be well tackled, the existing works on XP have only succeeded in assuming the particle orientation and guessing the canting angle. To this end, attempt is made in this research work to propose a model for calculating the actual canting angle by investigating the forces influencing orientation of the falling particle and the use of reliable measure of turbulence shear and inertial torque. Furthermore, mathematical models of dust induced complex scattering are examined by considering dust particle shapes and best fit ellipsoids. The scattering coefficients are derived and expressed in forms of attenuation and phase rotation using Rayleigh method. Three conditions are set to validate the suitability of the Rayleigh approximation for the model. It is shown that the method is valid for determining the scattering of spherical and ellipsoidal dust particles for small particle sizes (much smaller than the wavelength of interest). Results of the proposed model compared with some published results show close agreement. Attenuation in dry dust is only significant when the visibility becomes severe or at low millimetre wave bands. Differential attenuation and phase rotation are then computed from the scattering model results. Lastly, cross polarization discrimination (XPD) being the parameter for characterization of XP is predicted using the differentials and the canting angles as inputs. The circular and 450 linear cross polarization values in dry dust were found to be 24.8 dB and 46.2 dB respectively for visibility of 0.1 km and 1 km at frequency of 50 GHz and 1 km path length. Results show that XP is significant during severe visibilities and for dry dust storm; the XPD is fairly good and acceptable for dual polarization systems.