When moving around, mobile phones in stand-by mode periodically send data about their positions. The aim of this paper is to evaluate how personal radiofrequency electromagnetic field (RF-EMF) measurements are affected by such location updates. Exposure from a mobile phone handset (uplink) was measured during commuting by using a randomized cross-over study with three different scenarios: disabled mobile phone (reference), an activated dual-band phone and a quad-band phone. In the reference scenario, uplink exposure was highest during train rides (1.19 mW/m2) and lowest during car rides in rural areas (0.001 mW/m2). In public transports, the impact of one’s own mobile phone on personal RF-EMF measurements was not observable because of high background uplink radiation from other people’s mobile phone. In a car, uplink exposure with an activated phone was orders of magnitude higher compared with the reference scenario. This study demonstrates that personal RF-EMF exposure is affected by one’s own mobile phone in stand-by mode because of its regular location update. Further dosimetric studies should quantify the contribution of location updates to the total RF-EMF exposure in order to clarify whether the duration of mobile phone use, the most common exposure surrogate in the epidemiological RF-EMF research, is actually an adequate exposure proxy.
The applicability of personal exposure meters (PEM) has successfully been demonstrated in several epidemiological studies to characterize personal exposure to environmental radiofrequency electromagnetic field (RF-EMF) such as mobile phone base stations or broadcast transmitters.1, 2, 3, 4, 5, 6 It is acknowledged, however, that personal measurements are affected by one’s own mobile phone use (uplink emissions), which is a severe limitation for the interpretation if one is interested in differentiating between exposure from one’s own mobile phone and other people’s mobile phone. Such a differentiation is important, as exposure of the body depends heavily on the distance to the source, which is different for one’s own mobile phone compared with other people’s mobile phone.
Mobile phones are emitting RF-EMF not only when being used for calls and texting, but also in the stand-by mode owing to its location updates; that is, changing from one cluster of base stations to the next.7 As a network is divided into cells (location areas), covered by a group of base stations, a mobile phone informs the cellular network about changes in its location area, based on different location area codes. Such location updates are necessary to maintain constant connectivity with the network. In particular, when moving in a car or train, a mobile device periodically sends information about its position while changing location. However, little is known so far on the extent of such location updates in real-life situations.
Most personal exposure assessment studies have focussed on environmental EMF, and thus exposure from one’s own mobile phone (uplink) is not of interest, and different strategies have been used to deal with that problem:8 (1) noting wireless calls in a diary and excluding the corresponding PEM measurements from the data analysis9 or (2) hiring people for taking measurements and forcing them to shut down their own mobile phones.8 The latter approach is the best solution from a scientific point of view, but is unlikely to be acceptable for volunteers of a population survey.1
With the diary approach,9 higher mean mobile phone uplink exposure levels for study participants owning a mobile phone compared with participants not owning a mobile phone (0.0417 vs 0.0189 mW/m2, respectively) had been observed.9 This difference may be explained by forgotten or imprecise diary entries, by difference in the behaviour between the two groups in terms of spending time close to other mobile phone users or owing to location update procedures of one’s own mobile phone in stand-by mode.
To systematically evaluate the impact of one’s own mobile phone in stand-by mode on PEM measurements, two measurement studies were conducted: a public transport study and a car study. As we hypothesized that the impact of one’s own mobile phone is increasing with increasing movement velocity, we included measurements from different types of settings: in trains, buses and cars, while moving and staying at train and bus stations. We also considered the frequency bands, Global System for Mobile Communications (GSM) and Universal Mobile Telecommunications System (UMTS) separately, as well as the distance between the mobile phone and the PEM.
We used a three-way randomized cross-over study design with three scenarios: (i) a disabled mobile phone (reference); (ii) a dual-band mobile phone (Nokia 2600) working on two frequency ranges: GSM900 (880−915 MHz) and GSM1800 (1710–1785 MHz); and a quad-band smart phone (Blackberry bold 8800 and an iPhone 4) capable of transmitting and receiving on four frequency ranges: GSM900, GSM1800, CDMA (Code Division Multiple Access, 850–1910 MHz) and UMTS (1920–1980 MHz).
An overview of the study design is shown in Figure 1. During the nonreference scenarios, the mobile phones were in stand-by mode without own use. Measurements were taken close to the mobile phone (proximal), with a distance of ∼10−15 cm between the PEM and mobile phone (for both studies), and distal from the source (exact location, see Figure 2), with a distance of ∼50 cm for the public transport study and ∼70–80 cm for the car study. During data collection in the framework of the public transport study, the device distant to the source was carried in a waist pack in front of the body (Figure 2) in order to maintain a distance of ∼50 cm from the emitting device.
The public transport study was carried out in four different settings: bus stop, train station, bus ride and train ride. Data collection took place during 3 weeks (from 25 January 2010 to 23 March 2010) in the morning and in the evening during regular commuting hours, always at the same time of the day and on the same travel routes. The scenarios were rotated each day to obtain for each scenario one morning and one evening measurement for each workday. The scenarios were rotated each day to obtain for each scenario one measurement for each workday. During the measurements, a prespecified activity diary was filled in to unequivocally attribute each measurement to the correct setting or area.
The car study consisted of five car rides, which were conducted on five different days between 13 November 2010 and 4 January 2011 on the same routes. In each ride, a distance of ∼280 km had been covered. Rural, urban and highway areas were defined when leaving or entering the city or a highway, respectively, by-passing the road sign. By using GPS (Global Positioning System) recordings, measurements of each ride were classified as rural, urban or highway measurements.
All measurements of both studies had been collected by the same trained collaborator.
We used two PEMs of the type EME Spy 120 (SATIMO, Courtaboeuf, France, http://www.satimo.fr/), which were placed proximal and distal to the mobile phone. This portable device is capable of measuring 12 different frequency bands of RF-EMF, ranging from 88 MHz (frequency modulation) to 2500 MHz (W-LAN). Uplink and downlink mobile phone bands are measured separately. The measured frequency ranges for the uplink bands are 880–915 MHz (GSM900), 1710–1785 MHz (GSM1800) and 1920–1980 MHz (UMTS), which fits to the emission spectrum of the used mobile phones. Note that CDMA is not in use in the study country.
The measurement interval was set to 4 s in order to collect a large amount of data points.
To take into account the measurements below the detection limit, arithmetic mean values, and other summary statistical measures were calculated using the robust regression on order statistics method10 for each setting on each day separately. If less than three measurements were above the detection limit for a given setting and frequency band, the arithmetic mean value was set to 0.000265 mW/m2.
A total of 109,668 measurements had been collected (64,551 measurements from the public transport study and 45,117 from the car study). The power flux density of the total uplink measurements of the three uplink bands combined (GSM900, GSM1800 and UMTS) was highly variable. For the reference scenario, highest uplink values were found during train rides (1.19 mW/m2), whereas lowest values occurred during car rides in rural areas (0.0012 mW/m2) (Figure 3a). Uplink levels during the reference scenario (mobile phone turned off) were higher in the public transport study than in the car study, and total uplink exposure mainly originates from GSM900 and GSM1800 frequency bands, although the contribution of UMTS is negligible (<0.001 mW/m2, except for train rides: 0.0013 mW/m2) (Figure 3a and 3b). Even during the quad-band scenario, the GSM bands were higher than the UMTS bands in all settings.
Public Transport Study
Total power flux density of all the measured frequency bands (88–2500 MHz) for all settings combined for the PEM placed proximal to the mobile phone was 0.65 mW/m2 in the reference scenario, 0.43 mW/m2 in the dual-band scenario and 0.73 mW/m2 in the quad-band scenario. The average proportions of uplink measurements in all four transportation modes combined were 81.6% (reference), 72.6% (dual-band) and 55.3% (quad-band), respectively.
For all settings and scenarios combined, the percentage of nondetects for the device in vicinity to the source was 60.8% (67.7% for the distant device) for GSM900 and GSM1800 combined and 98.2% for UMTS.
During the scenarios with activated phones, GSM uplink (combination of GSM900 and GSM1800) measurements in public transports were not consistently higher compared with the reference scenario (Figure 3a and 3c), as would have been expected on the basis of our hypothesis. During train rides, where most of the location updates are expected to occur, measurement levels were actually lower with the activated phones. In contrast, UMTS uplink levels were always higher in the scenario with an activated quad-band phone compared with the two other scenarios without own UMTS emissions (Figure 3b). Except during train rides, this difference was smaller for the distal measurement device (Figure 3d). The data distribution for each scenario and frequency band is presented as Supplementary Figure 1a–d.
Total power flux density of all measured frequency bands in all areas for the PEM placed proximal to the mobile phone was 0.12 mW/m2 in the reference scenario, 0.35 mW/m2 in the dual-band scenario and 1.62 mW/m2 in the quad-band scenario. The proportions of uplink bands were 4.9%, 62% and 81.9%, respectively. For all settings and scenarios combined, the percentage of nondetects for the device in vicinity to the source was 88.4% (93.4% for the distant device) for GSM900 and GSM1800 combined and 99.2% for UMTS.
During the scenarios with activated phones GSM uplink measurements were considerably higher compared with the reference scenario (Figure 3a–d). For instance, in rural areas, GSM uplink of the proximal device was 0.0014 mW/m2 for the reference scenario, 0.395 mW/m2 for the dual-band scenario and 2.923 mW/m2 for the quad-band scenario (Figure 3a). The proximal and distal devices showed similar values for GSM frequency bands.
With regard to UMTS uplink, levels were increased for the quad-band scenario compared with the two other scenarios (Figure 3b). This increase was more pronounced for the proximal device than for the distal device. For the distal device, it was even negligible for the urban areas (Figure 3d). The data distribution for each scenario and frequency band is presented as Supplementary Figure 1a–d.
Our study demonstrates that PEM measurements are affected by one’s own mobile phone in stand-by mode. The effect was more pronounced in the car study than in the public transport study. This pattern is not surprising because measurements in one’s own car are hardly affected by other people’s mobile phone. During commuting in public transports, however, other people’s mobile phones are influencing the uplink measurements considerably. Thus, GSM levels in the reference scenario during bus and train rides were about 100 times higher than those during car rides. As a consequence of this high background exposure in trains, due to the use of other people’s mobile phone in a closed area intensified by the Faraday cage effect, the relative contribution of the location update from one’s own mobile phone is small and the contribution of the own mobile phone is masked in our measurements.
This measurement study provided additional insights. First, UMTS uplink exposure is considerably lower than GSM uplink exposure. For UMTS, the impact of the own quad-band mobile phone (smart phones) was observable in almost all scenarios. However, the absolute contribution of UMTS signals to total uplink exposure (GSM900, GSM1800 and UMTS signals combined) was very small (0.2% for the public transport study and 5.4% for the car study). Second, for location updates, quad-band phones seem to use both the GSM and the UMTS frequency bands. We measured higher GSM than UMTS levels and found an indication that GSM location update of quad-band phones is more pronounced than GSM location update of dual-band phones. This suggests that quad-band phones execute more location updates than dual-band phones. Possibly, quad-band mobile phones need more frequent location updates because of new applications (apps) including push notifications. Push notifications, which require W-LAN or cellular connection, are a way for applications (newspaper, e-mail, messages and others) to provide alerts and information. Third, even for the distal PEM of the car study, we found considerable impact from one’s own mobile phone. This implies that one’s own mobile phone in a car is a relevant exposure source to the passenger(s) even if not carried on the body.
Our study implies that PEM uplink measurements are affected by one’s own mobile phone in stand-by mode. This was best visible in the car study, where measurements were barely affected by other people’s mobile phone. In public transports or when being stationary (in bus stop and train station), the relative impact of one’s own phone was small compared with the other sources, and thus less clearly visible. Nevertheless, an impact on the measurement has to be expected, especially when moving. We also found some indications that RF-EMF contributions in stand-by mode will become more relevant in the future because of the increasing use of smart phones that need regular location updates. To the best of our knowledge, no study has yet investigated the exposure from location updates in real-life situations.
The study offers amendatory information about exposure provoked by other people. In this context, we observed higher RF-EMF exposure in settings in which a lot of people are present, as especially was perceived in public transports, particularly in trains, and in urban environments augmenting exposure levels. This was clearly shown by higher exposure levels for the reference scenario in which only background exposure levels were quantified.
Our results reflect a snapshot in time based on one type of mobile phone for each scenario and two mobile phone operators. Thus, it cannot be generalized to other countries or to the future, as the extent of location update is determined by various factors such as the type of phone and the implemented technology of the mobile phone network operators.11 Thus, there is an urgent need to evaluate more thoroughly how personal RF-EMF exposure is affected by one’s own phone in stand-by mode. A better knowledge of the relevance of this exposure source in comparison with RF-EMF exposure when talking on a phone helps clarify whether the duration of mobile phone use, the most common exposure surrogate in epidemiological RF-EMF research, is actually an adequate exposure proxy. In particular, when interested in whole-body exposure, new exposure assessment approaches have to be considered by taking into account the emission behaviour of mobile phones in stand-by mode. Whole-body exposure is of interest, for instance, in studies on leukaemia12 or on the foetus during pregnancy.13, 14
Our measurement study has some relevance for people who want to minimize their personal exposure. The study indicates that own uplink exposure during car driving can be considerably reduced (about a fraction of 100) when turning off one’s own mobile phone in order to prevent it from location updates. Recently, the use of UMTS phones has been recommended as a precautionary measure, because UMTS calls are carried out with lower amount of radiation emissions.15 Before this precautionary measure can be firmly given to the public, it has to be ensured that lower exposure during calls is not compensated with higher emissions in stand-by mode.
In summary, this study demonstrates the complexity of the RF-EMF emission pattern of mobile phones in stand-by mode. So far, this exposure source has been neglected in the RF-EMF research. More thorough studies are needed to quantify this contribution to the total personal exposure. Such knowledge is needed for the interpretation of previous RF-EMF research and for the design of future high-quality epidemiological research.
Röösli M, Frei P, Bolte J, Neubauer G, Cardis E, Feychting M et al Conduct of a personal radiofrequency electromagnetic field measurement study: proposed study protocol. Environ Health 2010; 9 (1): 23.
Viel JF, Cardis E, Moissonnier M, de Seze R, Hours M . Radiofrequency exposure in the French general population: band, time, location and activity variability. Environ Int 2009; 35 (8): 1150–1154.
Bolte JF, van der Zande G, Kamer J . Calibration and uncertainties in personal exposure measurements of radiofrequency electromagnetic fields. Bioelectromagnetics 2011; 32 (8): 652–653.
Thuróczy G, Molnar F, Jánossy G, Nagy N, Kubinyi G, Bakos J et al Personal RF exposimetry in urban area. Ann Telecommun 2008; 63 (1-2): 87–96.
Joseph W, Vermeeren G, Verloock L, Heredia MM, Martens L . Characterization of personal RF electromagnetic field exposure and actual absorption for the general public. Health Phys 2008; 95 (3): 317–330.
Frei P, Mohler E, Bürgi A, Fröhlich J, Neubauer G, Braun-Fahrländer C et al Classification of personal exposure to radio frequency electromagnetic fields (RF-EMF) for epidemiological research: Evaluation of different exposure assessment methods. Environ Int 2010; 36 (7): 714–720.
Lin YB, Lee PC, Chlamtac I . Dynamic periodic location update in mobile networks. IEEE Trans Vehicular Technol 2002; 51: 615–624.
Joseph W, Frei P, Röösli M, Thuróczy G, Gajsek P, Trcek T et al Comparison of personal radio frequency electromagnetic field exposure in different urban areas across Europe. Environ Res 2010; 110 (7): 658–663.
Frei P, Mohler E, Neubauer G, Theis G, Bürgi A, Fröhlich J et al Temporal and spatial variability of personal exposure to radio frequency electromagnetic fields. Environ Res 2009; 109 (6): 779–785.
Röösli M, Frei P, Mohler E, Braun-Fahrländer C, Bürgi A, Fröhlich J et al Statistical analysis of personal radiofrequency electromagnetic field measurements with nondetects. Bioelectromagnetics 2008; 29 (6): 471–478.
Vrijheid M, Mann S, Vecchia P, Wiart J, Taki M, Ardoino L et al Determinants of mobile phone output power in a multinational study: implications for exposure assessment. Occup Environ Med 2009; 66 (10): 664–71.
Cooke R, Laing S, Swerdlow AJ . A case–control study of risk of leukaemia in relation to mobile phone use. Br J Cancer 2010; 103 (11): 1729–1735.
Divan HA, Kheifets L, Obel C, Olsen J . Prenatal and postnatal exposure to cell phone use and behavioral problems in children. Epidemiology 2008; 19 (4): 523–529.
Vrijheid M, Martinez D, Forns J, Guxens M, Julvez J, Ferrer M et al Prenatal exposure to cell phone use and neurodevelopment at 14 months. Epidemiology 2010; 21 (2): 259–262.
Baumann J, Landstorfer FM, Geisbusch L, Georg R . Evaluation of radiation exposure by UMTS mobile phones. Electron Lett 2006; 42 (4): 225–226.
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Journal of Exposure Science and Environmental Epidemiology website
About this article
Cite this article
Urbinello, D., Röösli, M. Impact of one’s own mobile phone in stand-by mode on personal radiofrequency electromagnetic field exposure. J Expo Sci Environ Epidemiol 23, 545–548 (2013). https://doi.org/10.1038/jes.2012.97
- radiofrequency electromagnetic fields (RF-EMF)
- personal exposure meters (PEM)
- location update
- mobile phone
APPLICATION OF TIME-AVERAGED AND INTEGRAL-BASED MEASURE FOR MEASUREMENT RESULTS VARIABILITY REDUCTION IN GSM/DCS/UMTS SYSTEMS
Radiation Protection Dosimetry (2019)
Commentary regarding “Factors associated with mental health among high school students in Iran: Does mobile phone overuse associate with poor mental health?”
Journal of Child and Adolescent Psychiatric Nursing (2019)
Not in My Neighborhood: A User Equipment Perspective of Cellular Planning Under Restrictive EMF Limits
IEEE Access (2019)
Effects of exposure to 2100 MHz GSM-like radiofrequency electromagnetic field on auditory system of rats
Brazilian Journal of Otorhinolaryngology (2018)
A histopathological and biochemical evaluation of oxidative injury in the sciatic nerves of male rats exposed to a continuous 900-megahertz electromagnetic field throughout all periods of adolescence
Journal of Chemical Neuroanatomy (2018)