„Entwicklung von Mess-und Berechnungsverfahren zur ... - BMU

Seite 8 von 187 des Abschlussberichtes "Entwicklung geeigneter Mess- und Berechnungsverfahren"
sight scenarios inside buildings), logper antennas with higher directivity have theoretical
drawbacks. Biconical or dipole antennas with low directivity do not exhibit this principal
drawback, but may offer a strong influence of the user as well as walls or furnishings onto the
results. These influences may be so strong, that logper antennas are highly recommended for
this type of measurements.
The measurements shall be performed in the far field of the cellular station. In regions outside
the safety distance a measurement of the electric field strength alone is regarded to be sufficient.
Electric and magnetic field components are coupled over the free space impedance at
much smaller distances from the antenna, as is defined by the far field formula based on the
geometrical size of the base station antenna.
Measurements with spectrum analyser or measurement receiver desire broad knowledge of
the basic operation of these devices as well as their settings. Especially resolution bandwidth,
detector type, frequency range, video band width and sweep time must be correctly matched
onto the base station signal to be analysed. Otherwise significant errors may be possible.
Unfortunately default settings are not suitable often for the present problem. Therefore the
most important settings are defined and summarised in a table.
Besides the measurement parameter settings also the measurement procedure has a large
influence on the measurement results. Here first of all the question has to be discussed, onto
whether the exposure maximum, or an averaged value over a volume has to be determined.
With sweeping method, rotation method and raster method three practically tested methods
are available. All methods lead to the same results, if performed carefully and under identical
conditions. Concerning time effort the raster method is most time consuming; the sweeping
method is the fastest one. With rotation method and raster method space problems may occur.
The spread of the results concerning measurements by different laboratories is comparable.
Especially the spread of the expensive raster method is not better than for the easier sweeping
method.
Here it must be taken into account, that the raster method alone provides an averaging inside a
measurement volume. A thorough investigation, however, shows, that an “optimal” average
geometry strongly depends on exposure scenario and no “general” geometry can be found. By
reducing the number of raster points to practical values the sensitivity of the averaging result
to exposure scenario and type as well as location of the averaging geometry rises. Therefore
the sweeping method is regarded to be the best suited for the present measurement problem.
Reproducibility of the sweeping method at successive measurements tends to be very good,
assuming a very carefully operation of the sweeping procedure to correctly capture all polarisations
and wave propagation directions in the volume.
The post processing covers the computation of the originally measured power or voltage
values into flux densities or field strength values, directly comparable with the exposure
standard limits. Also, the measured momentary exposures must be extrapolated to the maximal
operational state of the station. For that, at GSM stations the exposure due to the time
constant BCCH channels is extracted and connected with the maximal possible number of
channels of the station. At UMTS, two trends can be observed at the moment: Here also
signalling channels do exist. These can be captured with a frequency selective measurement
and extrapolated onto the maximal state under the worst case assumption, that only signalling