Thielens Arno, Agneessens Sam, Verloock Leen, Tanghe Emmeric, Rogier Hendrik, Martens Luc, Joseph Wout
Wireless and Cable Group, Department of Information Technology, Ghent University/iMinds, Gaston Crommenlaan 8 Box 201, Ghent B-9050, Belgium
Electromagnetics Group, Department of Information Technology, Ghent University, Sint-Pietersnieuwstraat 41, Gent B-9000, Belgium.
Radiat Prot Dosimetry. 2015 Jan;163(1):58-69. doi: 10.1093/rpd/ncu056. Epub 2014 Apr 11.
Two radio-frequency personal exposimeters (PEMs) worn on both hips are calibrated on a subject in an anechoic chamber. The PEMs' response and crosstalk are determined for realistically polarised incident electric fields using this calibration. The 50 % confidence interval of the PEMs' response is reduced (2.6 dB on average) when averaged over both PEMs. A significant crosstalk (up to a ratio of 1.2) is measured, indicating that PEM measurements can be obfuscated by crosstalk. Simultaneous measurements with two PEMs are carried out in Ghent, Belgium. The highest exposure is measured for Global System for Mobile Communication downlink (0.052 mW m(-2) on average), while the lowest exposure is found for Universal Mobile Telecommunications System uplink (0.061 μW m(-2) on average). The authors recommend the use of a combination of multiple PEMs and, considering the multivariate data, to provide the mean vector and the covariance matrix next to the commonly listed univariate summary statistics, in future PEM studies.
将两个佩戴在双髋部的射频个人暴露计（PEM）在消声室内的一名受试者身上进行校准。利用该校准确定PEM对实际极化入射电场的响应和串扰。当对两个PEM的响应进行平均时，PEM响应的50%置信区间减小（平均减小2.6 dB）。测量到显著的串扰（高达1.2的比率），表明PEM测量可能会因串扰而混淆。在比利时根特使用两个PEM进行了同步测量。测量到全球移动通信系统下行链路的暴露最高（平均为0.052 mW m(-2)），而通用移动通信系统上行链路的暴露最低（平均为0.061 μW m(-2)）。作者建议在未来的PEM研究中，除了通常列出的单变量汇总统计数据外，结合使用多个PEM，并考虑多变量数据，提供均值向量和协方差矩阵。
