Statistical analysis of accurate prediction of local atmospheric optical attenuation with a new model according to weather together with beam wandering compensation system: a season-wise experimental investigation. Issue 13 (19th July 2016)

• Record Type: Journal Article
• Authors: Arockia Bazil Raj, A.; Padmavathi, S.
• Is Part Of: Journal of modern optics. Volume 63:Issue 13(2016)
• Page Start: 1286
• Page End: 1296
• Publication Date: 2016-07-19
• Subjects: Meteorological data -- atmospheric attenuation -- beam wandering mitigation -- regressive model -- validation test -- visibility and seasonal investigation; Optics -- Periodicals; 535
• DOI: 10.1080/09500340.2016.1140840
• Languages: English
• ISSNs: 0950-0340

Abstract

Abstract: Atmospheric parameters strongly affect the performance of Free Space Optical Communication (FSOC) system when the optical wave is propagating through the inhomogeneous turbulent medium. Developing a model to get an accurate prediction of optical attenuation according to meteorological parameters becomes significant to understand the behaviour of FSOC channel during different seasons. A dedicated free space optical link experimental set-up is developed for the range of 0.5 km at an altitude of 15.25 m. The diurnal profile of received power and corresponding meteorological parameters are continuously measured using the developed optoelectronic assembly and weather station, respectively, and stored in a data logging computer. Measured meteorological parameters (as input factors) and optical attenuation (as response factor) of size [177147 × 4] are used for linear regression analysis and to design the mathematical model that is more suitable to predict the atmospheric optical attenuation at our test field. A model that exhibits the R 2 value of 98.76% and average percentage deviation of 1.59% is considered for practical implementation. The prediction accuracy of the proposed model is investigated along with the comparative results obtained from some of the existing models in terms of Root Mean Square Error (RMSE) during different local seasons in one-year period. The average RMSE value of 0.043-dB/km is obtained in the longer range dynamic of meteorological parameters variations.