Walter Vollenweider, Siemens Schweiz AG

It is usually assumed that the pathloss can be expressed as:
A [dB] = A0 + n ? log d
where n is a value between 20 dB/decade in free space and 40 dB/decade if the ground is an ideal conductor. This model seems to be quite useful over open terrain. It shows deficiencies in a built up area, and it is quite useless in a building. We found that the expression
A [dB] = A0 + n ? log d + m ? d
where n is usually 20 dB/decade and m is between 1 and 2 dB/m gives much better results. Fading is generally assumed to be either of the Rayleigh or the Rice type, which is certainly true outdoors. Measurements in buildings did not agree very well with these models. It would be possible to use the Nakagami model which agrees better with our results. The underlying assumptions can, however, not be applied to the situation in a building. It seems that the fading can be more accurately approximated by a probability density function which increases linearly by a value which is close to 10 dB/decade.
Measurement results from some buildings are shown and compared to the expected values.