Wireless channel characterization is of critical importance in providing reliable tools for link budget prediction and estimation of wireless propagation, particularly in the case of an obstacle-dense environment. Based on measurements performed in a series of indoor propagation topologies, a novel empirical method for the calculation of shadow depth has been implemented and evaluated. This work presents further investigation into the log-normal nature of large-scale fading, by examining the variations of the measured values of local-mean received power throughout these topologies. The impact of these variations on the estimation of Outage Probability is shown, and the realistic deviations from the theoretical log-normal scenario are discussed. Furthermore, a shadow-based approach for the calculation of the attenuation over distance is presented and evaluated compared to the Devasirvatham model. It should be noted that whereas our work focuses on the 2.4 GHz channel, this methodology can be applied to any frequency of interest. Results confirm the importance of this novel methodology concerning wireless channel characterization for obstacle-dense indoor propagation topologies.