Answers are grounded in this paper and cite their sources.
Time delay spread and signal level measurements were made in two dissimilar office buildings at 850 MHz, 1.7 GHz and 4.0 GHz. The RMS time delay spread did not deteriorate with increasing frequency in these buildings. Its maximum value at the three frequencies did not exceed 270 ns at the larger building and 100 ns at the smaller building. Attenuation was nearly the same at the three frequencies, but decreased slightly with increasing frequency, within the limits of the experiment's accuracy.
No takes yet. Share an insight, caveat, or question.
Devasirvatham et al. (1990) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: