home
page
brief
description
exposure
guidelines
legal
frame
equipment and measurement procedures measurement
results

 

EQUIPMENT

Á. Antennas

Biconical antenna

EMCO(93110)

Biconical antenna

(SCHWARZBECK UBAA 9115)

Biconical antenna

(SCHWARZBECK SBA 9113)

Receiving antennas with an isotropic radiation pattern in at least one plane and linear polarization were used (like the pictures above). In order to estimate the total power density from a radiation source, three measurements need to be done: two of them in horizontal polarization with axes vertical to each other and one in vertical polarization.

Â. Receiver

Spectrum Analyser (Hewlett-Packard 8593ÅÌ)

The spectrum analyzer which has been used can be connected to a computer. The communication takes place via the GPIB port and through the IEEE-488.2 protocol. Computer aided controlling, versus to manually performed measurements, has the advantage of allowing the automation of the whole process, while decreasing the possibility of human faults or omissions. Moreover, the collection of  more data in less time becomes possible.

C. Other
  • Connection cable RG-8/U, (Belden 9913). It introduces low losses along the whole measured spectrum and also very good shielding, so as to prevent the reception of noise signals.

  • GPS device (GARMIN eTrex Summit)

  • Digital camera (300 DIGITAL, Olympus)

 

MEASUREMENT PROCEDURE

All measurements took place outdoors, near GSM base stations (50-250m distance), that can be freely accessible by the general population (e.g. roads, playgrounds, schools, squares, or the vicinity of public buildings). The measurement points were chosen in such a way that they are distributed along the whole urban area of Thessaloniki. In the municipalities of the rest north Greece, the measurement places were selected, in order to be near to places of public access and interest (schools, squares, central streets, e.t.c.). 

The frequency range which was measured in the present study (20-3000MHz) was divided in a number of frequency bands. An appropriate antenna was used for each band. The antennas were calibrated for the frequency range for which they were used. The measuring antenna was placed at a height between 1.9 and 2.2 meters above the ground. Each of the three frequency ranges which were measured with a specific antenna type, was divided into subbands. The spectrum analyzer was appropriately initialized, in order to guarantee: 

  • sufficient distinction between the signals of each subband

  • accuracy in the estimation of the total power density, by choosing an intermediate frequency filter proportional to the expected frequency range of the measured source, where possible, and

  • measurement time that allows completing a measurement of the total frequency range within an acceptable time.

The estimation of the power density and the corresponding SI in each frequency subband (e.g. FM, GSM) is made by processing spectra that are received with different intermediate filters. The identification and the detection of the peak values (of the emissions) is made in the spectrum recording taken with the narrowest intermediate filter. On the other hand, the estimation of the final value is made with the values taken with a wide intermediate filter, in order to overestimate the power corresponding to each discrete peak (emission frequency).

By using a software of automated measurement, there is the possibility to divide each frequency range into subbands, in order to achieve a detailed scan. Each frequency subrange can be scanned up to 50 times.The data are transfered and saved in the computer's hard disk, for further processing.

It is worth noting that the temperature and humidity at the measurement points were not recorded.