Machine Learning Approach to QoS Adjustment Leveraging LCR in Mobile Networks Limited by Beaulieu-Xie Fading and Co-Channel Interference

Future applications of wireless systems requires as better channel models as possible. For this reason, a few years ago new fading model was discovered by Norman Beaulieu and Xie Jiandong for wireless systems containing multiple dominant specular components. This fading model is named after the authors Beaulieu-Xie fading and distribution is given in closed form. It has been shown that this model can be more useful than the known Ricean and Nakagami models for describing wireless environments with line-of-sight (LoS) and multiple reflected specular components (non-line-of-sight (NLoS) environment). In this work, we analyzed wireless system with selection combining in the presence of Beaulieu-Xie fading and co-channel interference. For such scenario, we derived the formula for level crossing rate (LCR) and plotted some graphs. Then, we examined the influence of the fading parameters, having real physical meaning and allow setting up of the fading severities. At the end, we make use of the calculated value of LCR for classification-based machine learning approach to Quality of Service (QoS) degree estimation, together with other factors (network load, temporal aspects).