You have no items in your shopping cart.
Abstract
Quality of service is one of the major factors demanded by customers in any communication network services. It is therefore essential to put into consideration seasonal propagation parameters before designing and establishing any wireless network systems in order to meet demands of customers. In order to achieve this, mathematical predictions and estimation of effects of change in season on GSM signal strength are needed for designing of wireless systems to ensure all year round effective services. In this thesis, investigation of RF channel behaviour based on extensive measurements of signal strength and other propagation parameters up to 1200 meters from the base stations adopting single sector verification technique in wet and dry seasons are presented. The measurements were carried out at 900 and 1800 MHz bands in urban, suburban and rural environments of Ondo, Ekiti and Osun States of Southwestern Nigeria using Sony Ericson TEMS phones and global positioning system connected to a laptop equipped with TEMS software and cell refs of the base stations in the studied areas. Seasonal two-slope path loss model was proposed for each environment based on the estimated break point distances and compared with the calculated path loss values alongside cost-231 Hata and three other models. However, for a single path loss model for each selected environment, the offset and slope parameters of cost-231 Hata model were adjusted with the measured path loss values and the mean seasonal path loss difference incorporated. Climatological parameters such as pressure, temperature and relative humidity at ground surface and at 100 meters altitude over all the monitored environments were obtained between 2016 and 2017. These data were used to compute the radio refractivity, refractivity gradient, k-factor and geoclimatic factor for each of the environment under consideration. The two-slope path loss models indicated a better prediction accuracy in terms of root mean square error, mean prediction error analysis and fitting accuracy; followed by cost-231 model when compared with other models. Mean path loss difference between wet and dry seasons was estimated using the developed two-slope model to be 2.5, 4.2 and 7.0 dB for urban, suburban and rural environments respectively. Adjusted cost-231 model exhibited improved prediction accuracy compared to the original cost-231 Hata model in terms of error analysis and fitting accuracy. Curve fitting of the measured path loss values with lognormal, weibul, extreme value, gamma, logistic and normal distribution models was performed. Results of three performance evaluations and curve fitting accuracy indicated that the normal distribution function exhibited a better curve fitting with the measured path loss values in both seasons. The estimated k factors and geoclimatic factors vary with months and season in each environment. Path loss values contour plots in both seasons were developed for all the selected environments. The overall results show that the adjusted model and normal distribution function are suitable for prediction and modelling of path loss in mobile micro cell coverage in the investigated region. Propagation parameters from this research will be useful in the design of wireless channel of micro wave bands in Southwestern Nigeria.
