Channel maps and stochastic models in elevation based on measurements in operating networks

In this contribution, a channel measurement campaign recently carried out in the Jiading campus of Tongji University, Shanghai, China, is introduced. The downlink signals of an active Universal Mobile Telecommunications System (UMTS) were collected at more than 4000 locations and processed to extract channel impulse responses (CIRs). The signal to noise power ratios (SNRs), delay spread, Doppler frequency spread, and the status of line-of-sight (LoS) existence are calculated and used to generate channel maps, i.e. geographic maps overlaid with spots at measurement locations being color-coded by the values of channel parameters. The variations of channel parameters with respect to the co-elevation in NodeBs are investigated. Relationships between the delay, Doppler frequency spreads and the LoS occurrence rate at specific co-elevations are analyzed. Based on these results, a new method is proposed which is applicable for generating realistic CIRs in time and frequency domains that satisfy the elevation-dependent statistics extracted from the measurements.