Telecommunication systems are presently at its lucrative growth-period in history; owing to supporting technologies and operational design that permit their wider deployment and acceptance. The need for better and robust telecommunication system networks design and management, estimating the strength of propagated field signal levels accurately at the user equipment terminal has become extremely important. In literature, one popular regression technique that is often engaged to perform field signal level predictive analysis and estimation of distributional strictures is the least square regression technique. This is probably due to its soft computational complexity and simplicity in graphical presentation procedure. Nonetheless, the resultant regression model may perform poorly owing to high stochastic nature and unequal noise variance (heteroskedastic) problems of most radio signal data. In this contribution, we proposed the application of weighted least square estimation approach combined with Poynting vector theory to model and estimate practical electric field strength data. The field strength data was obtained over radio frequency system interface which belongs to a commercial LTE cellular broadband networks service provider operating in typical urban area. In terms of reliability and precision, the outcome show that the employed hybrid weighted least square has the best performance compared to using the standard least square method. We also show that the precision performance improves as the power of the weighted least square method grows.