Calibration of Implantable Electric Field Probes for Mobile Phone Dosimetry

摘要

Widespread use of mobile phones raises important health issues which are of concern to the general public. To guarantee good performance the mobile handset needs to radiate few hundred milliwatts in order to establish a reliable communication link with the base station some distance away. But with a radiation source just few millimetres away from human head it is not surprising that researchers have discovered that over 40percent of the RF power transmitted may be absorbed by the mobile phone user [1,2,3]. For this reason the monitoring of radiated power and the power absorbed by the user is important. The quantity of interest is usually the Specific Absorption Rate (SAR), the power deposited in a given mass of body tissue, expressed in W kg~(-1). It has been found, however, that one of the best practical methods for estimating SAR is by the use of miniature isotropic 'implantable' RF electric field ('E-field') probes. A number of these E-field probes are available on the market. The probes are generally used to measure RF fields in tissue-equivalent phantoms. In the case of mobile phone safety testing the phantoms consist of artificial heads and hands exhibiting various degrees of geometrical approximation to their biological counterparts, but having dielectric properties close to those of the body tissues. The phantom heads generally contain brain-shaped cavities which can be filled with a liquid which simulates the dielectric properties of brain tissue. The probes can be inserted and scanned in this cavity to measure the radiation dosage from cell phones. This paper is concerned with the calibration and characterisation of these implantable probes, the main difficulty being that they must be calibrated for measurement of field strength in the same dielectric liquid as is used in the phantoms. The reason for this is that the radiation wavelength and impedance-match change with the permittivity of the medium and so the calibration factor of the probes is significantly affected by the transmission medium. Two frequency ranges have been of concern - close to 900 MHz and close to 1.8 GHz - but requirements to expand to frequencies close to 410 MHz and above 2 GHz are already clear as cell phone technology continues to expand. In order to provide such measurements with reliable traceability, NPL has been engaging in research covering all factors governing probe performance including permittivity and E-field measurement, cell-phone communication protocols, locating the phase centre of the probe antennas and proximity effects. The work reported here covers the first NPL prototype system which operates at frequencies near 1.8 GHz. The 1.8 GHz system uses IEC R-22 (British WG-8) waveguide as the calibration cell, see Figure 1. This is large enough to allow access for typical probes via a slot in the centre of the broad face of the waveguide whilst avoiding excessive probe/cell interaction effects, though these cannot be entirely avoided - see below. Typical probes are 5 - 12 mm in diameter and are connected to their detection system via high impedance leads in tubular polymer sheathes. The waveguide cell can be used for probe calibration both in air, for which calibrations are also available from the UK National Standards facility, and also in brain simulation liquid. The health and safety applications of these probes require that all three components of field strength be measured and this is usually realized by using three mutually orthogonal dipoles as detectors. This requires that determination of antenna pattern and isotropy must be undertaken for comprehensive characterization. Preliminary calibration results at 1.8 GHz are reported in this paper and a preliminary uncertainty budget has been drawn up. The advantages and shortcomings of this prototype method are discussed and further options for calibration are considered.