The uplink (from handset to satellite) frequency band is in L band and is stated to be around 1.6GHz.
It is possible that "smart antenna" technology will be needed to aim the directional antennas on the handset at the satellites.
In any event, it is likely that the output power from the mobile handset will need to be be significantly greater than that from ordinary terrestrial digital mobile phone technology, such as GSM. Figures of up to "7 watts of total radiated power" have been mentioned. It is not clear from the reports whether this figure represents a peak power level or an average over some time span which is not short compared to the thermal time constants in the brain. There is an argument made that it is the peak power levels which may be responsible for DNA damage in tissue, and for electronic processes in nerve fibre. Reports also suggest that the link budget will be marginal for most of the time, and that the technology may not work reliably indoors or under forest canopies.
In 2005, some measurements have been made on a standard GSM phone supplied by DST for the UK market, around Guildford. The reported results are :- at 1800MHz this phone was putting out a maximum of 1 watt of power (+30 dbm), on the majority of sites around the town including some with line-of-site to a base station mast, backing off in areas of high signal strength to +22 dbm. At 900 MHz it is said to put out a maximum of 2 watts (+33 dbm). This is not a 3G phone, so already we have significant transmitted power levels around a majority of the service area. Again, taking the phone into a fringe area causes the power control to ratchet up the handset transmit power to maximum. The output power level was also said to have been monitored in a UniS anechoic chamber, and the measurements apparently agreed with the reported output power.
With advances in battery technology, there is every incentive for phone manufacturers to keep increasing the transmit power levels from their handsets.
There was a report in the British newspapers, about the middle of March 1999, in which a heavy user of BT's mobile phone technology was said to have been suffering health problems and was trying to obtain a legal ruling that this might be in respect of his heavy use of a transmitting mobile phone in the course of his job. (Several hours a day of use were mentioned).
There is also a report in December 2004 on the BBC news site that mobile phone radiation alters the structure of DNA. This might be expected, for DNA is a long coiled springy molecule and may have resonant frequencies in the microwave region which would absorb energy from the fields. Your Author has always regarded this mechanism as the most likely one to cause problems, rather than the direct heating of absorbing tissue.
The current NHS guidelines are equivocal.
Microwave radiation is sometimes believed to cause cataracts in the eyes, possibly to cause brain tumours, and perhaps to cause memory loss in some susceptible people, at sufficiently high power levels. There is much ongoing debate in the community about the precise power levels which might cause these effects. In our laboratories, we warn people not to look into a transmitted microwave beam at a frequency of 10GHz and a total radiated power level of 50 milliwatts.
There is some epidemiological evidence from people who live reasonably close to fixed base station transmitters and have been exposed to quantifiable radiation levels over long periods of time, that at sufficiently low levels of exposure there is no measurable risk. Thus there is a lower limit on the field strengths which might be of concern. The question here is about the limit above which radiation is proven to be of concern.
Nevertheless, a peak power level of up to 7 watts at 1.6 GHz would appear to be cause for concern. The epidemiological evidence for any potential health hazards would not be available for 25 years, or a human generation, to allow the extra incidence of medical effects to be made manifest in the community.
The sole purpose of this web page is to point out that in a contentious area like this safety issue, independent scientific evidence is hard to come by. At present there does not seem to be enough generally agreed data to decide whether or not there may be a problem. Here, we suggest that the commercial pressures may seek to minimise potential risks. Is it not important to put into place properly constructed and funded longitudinal epidemiological studies at the time this new service is commissioned?
There was a select committee of the UK House of Commons which investigated the issue of hazards from mobile phones. Their minutes took evidence from representatives of mobile phone companies having a commercial interest in purveying this technology. These gentlemen maintain that there is no additional hazard from the extra power levels needed for satellite phone handsets (see Questions 165-171) as the energy is primarily beamed upwards. This is a disingenuous answer, as "upwards" in their terms is defined as above an elevation of ten degrees above the horizon. They further state that the antennas are bulky and that they are held mostly above the head.
This argument shows a neglect of the near field properties of the antennas. The far field radiation pattern is directed into the upward hemisphere, but this says nothing about the radiation polar pattern in the near field, close to the handset. The antennas notes elsewhere on these pages point out that the radiation pattern is only independent of distance from the source once we have reached the far field.
You may be interested to know that one of my PhD students has run simulation studies comparing the absorption in the head between standard terrestrial set-ups (monopole on a box) with satellite situations (quadrifilar helix on a box). For a given radiated power the satellite set-up produces 1.4 times less peak absorption rate. If we rearrange the feed on the satellite case we can reduce it by a further 10dB or so.
-----------------------------------------------
The note above about the quadrifilar helix needs a little explanation. A helical antenna can be arranged to have "end fire" boresight characteristics, by choosing the diameter and pitch appropriately compared to a wavelength. A "quadrifilar" helical antenna consists of four intertwined helices. Imagine four strips of spaghetti, placed side by side in a square arrangement, and then twisted about their common length. If one takes a cross section at any point along the helix, at right angles to the winding axis, one sees an arrangement of conductors similar to an array of isotropes. If one makes this array to be a square of side 8.33cm or about one half-wavelength, then far field lateral radiation is minimised from the array factor. The side of the square can be reduced of the helices are encased in dielectric, but that will make the tuning more critical and the performance more susceptible to objects in the near field adjacent to the antenna.
The handset, when held beside the head, can be arranged
to fire upwards with minimal sideways radiation into the
head. Of course, this is only approximately true as the
head will be in the near field of the sideways radiation
from the antenna. Generally speaking, to construct
a small antenna which is directional, and suitable for use
at 1800MHz on a hand held communicator, is difficult.
Normally fields are calculated only in the far
field region. The NEC simulation code is not
necessarily accurate in the near field as it relies
on a specification of the source conduction currents
only. Also, the conditions of use of the handset
are not under the control of the antenna designer;
as in all "specificationmanship" the tenet of "caveat
emptor" applies.
-----------------------------------------------