Comparison of various safety guidelines for electronic article surveillance devices with pulsed magnetic fields

The paper uses the two methods suggested in both the ICNIRP and proposed IEEE safety guidelines for compliance testing of security systems based on time-varying magnetic fields being introduced for electronic article surveillance (EAS), radio-frequency identification, and other applications. For nonsinusoidal pulses that are often used, the two procedures are to treat the exposure as a multifrequency exposure with various frequency components or to calculate the peak induced current densities or electric fields treating the highest of the pulses of duration t_p as a single frequency, half sinusoid of the same duration and frequency 1/(2t_p). Using either of the procedures, the induced current densities (J) or electric fields (E) are compared to the basic restrictions on J or E for compliance testing. Using a heterogeneous, tissue-classified anatomic model of the human body, we calculate the induced J and E for the various tissues for a realistic, EAS system for two typical nonsinusoidal pulses to show that the two methods give substantially different results. While the approximate but simpler method of treating the pulse as a half sinusoid results in peak induced J or E that may be compliant with safety guidelines, the rigorous method of treating such exposures as multifrequency exposures gives induced current densities or electric fields that may be up to twice as large, thus making such systems potentially noncompliant with the safety guidelines. Since it is more accurate, it is suggested that safety assessment based on the Fourier analysis leading to multifrequency components be used for compliance testing of such devices.