Radiation from Cellphone
Transmitter Towers
in the United Kingdom
Colin Baxter
August 2003 (Updated March 2004)
A. Theoretical Predictions of Radiation Power Densities
1.
The theoretical calculation of radiation power densities in mobile
communication involves constructing a mathematical model.
The modelling process attempts to take into account local conditions:
such as the roughness of the ground; the presence of buildings,
trees and other obstacles; the topography of the area; etc.
However, no single model fits all conditions and locations -
and all models by their nature are approximations.
2.
The starting point in the modelling process is the so-called free-space
path loss formula
[1],
which gives the radiation strength at a distance from
the antenna in ideal conditions on a perfectly flat earth.
3.
The formula assumes that the heights of the antenna and radiation
target above the ground are negligible compared to the distance from
the antenna to the target, i.e. the point at which the radiation
power is to be determined.
This is a major flaw.
In the case where the assumption no longer applies then the
formula is invalid and the radiation power varies sharply over very
short distances (of the order of metres).
This spatial inhomogeneous nature of the radiation field means
that the power density is highly localised and there
exists no unique representative value.
In other words, radiation power predictions in the vicinity of
a transmitting mast are unreliable if they have been obtained
by standard modelling.
4.
In practice, the modelling of theoretical predictions of radiation
power proceeds along one of two ways.
The free space path-loss formula is modified either by use of an
algorithm to classify local conditions as one of a set of
generalised standard environments
[2] -
e.g. “desert”, “urban”, “open country” - or the geographical
details of the radio cell are overlaid on a topographical map of
the area and local conditions, such as the shape of the ground
contours, are averaged from the topographical data.
The latter method can be used in the UK, where appropriate data is
available from the Ordnance Survey.
5.
In either procedure, the accuracy of the predictions - in areas
where the free space path-loss formula is valid - depends on how
well the actual local conditions have been simulated and how well
such simulations can predict appropriate modifications to the free
space path-loss formula.
No indications of the degree of error associated with the
radiation-power prediction data are usually provided by the mobile
phone operators.
Instead, they often say that their calculations are “based on the
worst case scenario”.
Of course, such a term is meaningless until the exact conditions
associated with “the worst case scenario” algorithm are defined.
B. Practical Measurements of Radiation Power Densities
1.
In 2001 the
Radiocommunications Agency (RA)
began an audit of the radiation emitted from base stations sited on school
buildings in the UK.
The RA audit remains one of the few programmes of measurement - as opposed
to theoretical modelling - of the radiation associated with mobile
radio-communication base stations.
2.
On 29 December 2003, the Radiocommunications Agency was subsumed
into a new agency called the
Office of Communications (Ofcom).
The activities of the RA are referred by Ofcom as a "legacy", but it is
not clear whether the radiation audit is continuing at the present time.
3.
A feature of the results of the RA audit is the wide variation in the
measured radiation power densities within a site.
For example, within the confines of Brixham Community College,
Brixham, which was audited by the RA on 23 July 2001, the highest
measured ICNIRP public exposure level of radiation
[3]
was nearly ninety (90) times greater than the lowest recorded level.
Within the confines of Folkestone School for Girls, Folkestone,
which was audited by the RA on 23 November 2001, the highest
measured ICNIRP public exposure level was over four-thousand (4000)
times greater than the lowest recorded level.
Again, within the confines of the Everlasting Ministries
Nursery School, London, which was audited by the RA on 21 November 2002,
the highest measured ICNIRP public exposure level was
five-hundred-and-eighty (580) times greater than the lowest recorded level.
4.
The schools mentioned in Paragraph B3 have been chosen at random
from the list of RA audited schools.
They are given here as a representative sample; but they are
indicative of the errors and general arbitrariness associated with
radiation power levels.
The results in Paragraph B3 illustrate the dangers of assigning a
particular value of radiation power density as a definitive indicator
of the ICNIRP public exposure level within a location.
5.
The RA audit also reveals wide variations in the measured radiation
power densities between sites.
For example, the highest measured ICNIRP public exposure level at
Folkestone School for Girls, Folkestone (audited by the RA on 23 November 2001)
was nearly three-thousand (3000) times greater than the highest
measured ICNIRP public exposure level at Countesthorpe Community College,
Countesthorpe (audited by the RA on 15 February 2001); over seven-hundred
(700) times greater than the corresponding level for Carrickfergus Grammar
School, Carrickfergus (audited by the RA on 24 September 2001); but only
one-and-a-half (1.5) times greater than the corresponding level for
Brixham Community College, Brixham (audited by the RA on 23 July 2001).
The antennae associated with all these schools were identical
(Rohde and Schwarz, Model HE200.4050.3609.02).
6.
As in Paragraph B3, the schools given in Paragraph B5 are presented
as a representative sample only.
Obviously, it is more problematic to compare the results associated
with different schools than those within the confines of the same school.
Nevertheless, they illustrate again the wide range in values found
in practical, measurements of radiation power density.
C. Conclusions:
I
Radiation power density predictions are highly model dependant.
II
In situations where the height of the antenna is not negligible,
compared to the distance from the antenna to the target, the free
space path-loss formula is invalid.
The radiation power density becomes highly localised, with large,
spatially-dependant variations.
III
Practical measurements undertaken by the Radiocommunications Agency
reveal wide variations in the values of radiation power-densities in
real geographical sites.
The variations indicate the myriad of factors involved in
radiation propagation, and the near-impossibility of
assigning a particular value of radiation power density as a definitive
indicator of the ICNIRP public exposure level within a particular location.
References
Click each item to see it's
BibTeX entry.
[1] R. C. V. Macario, Cellular Radio: Principles and Design, Macmillan (London, 1997).
[2] William C. Y. Lee, Mobile Cellular Telecommunications Systems, McGraw-Hill (New York, 1989).
[3] Guidelines for Limiting Exposure to Time-Varying Electric, Magnetic and Electromagnetic Fields (up to 300 GHz.), International Commission on Non-Ionizing Radiation Protection (ICNIRP), Health Physics Society, 1998.
