Cell-phone interference with pocket dosimeters

Cell-phone interference with pocket dosimeters David Djajaputra, Ramasamy Nehru, Philip M. Bruch, Komanduri M. Ayyangar, Natarajan V. Raman, and Charles A. Enke Department of Radiation Oncology, University of Nebraska Medical Center, 987521 Nebraska Medical Center, Omaha, NE 68198-7521 Abstract. Accurate reporting of personal dose is required by regulation for hospital personnel that work with radioactive material. Pocket dosimeters are commonly used for monitoring this personal dose. We show that operating a cell phone in the vicinity of a pocket dosimeter can introduce large and erroneous readings of the dosimeter. This note reports a systematic study of this electromagnetic interference. We found that simple measures are enough to mitigate this problem such as limiting the distance between the cell phone and the dosimeter to greater than 30 cm or shielding the dosimeter, while maintaining its sensitivity to ionizing radiation, by placing it inside a common anti-static bag.

E-mail: [email protected]

Cell-phone interference with pocket dosimeters

2

1. Introduction Millions of cell phones are currently in use around the world. In the United States alone, there were over 140 million subscribers in the year 2002 (US Census Bureau 2002). This number has been forecast to grow to about 187 millions in year 2004 (www.eMarketer.com 2004). Worldwide, there were about 625 millions cell-phone users in year 2000 and this number was estimated to increase to 1.15 billion in year 2004 (Industry Standard 2000). Most of current cell phones use digital technology and operate at two low signal strengths: 0.6 and 3 W. Older analog models typically use higher transmission power, but these models are rapidly declining in number. Cellular telephony divides the coverage area into contiguous cells of about 10 square miles area (in the sub-urban; smaller area is used in urban area), each with a base station. In addition to conserving the battery power, the relatively low transmission power used by cell phone is required by the system to limit its signal outside the cell in which it is located. This cellular division principle allows the same frequency to be reused by different cell phones located in non-adjacent cells and therefore increases the number of users that can be served. Although the microwaves emitted by cell phones are of relatively low power, it is still capable of interfering with the operation of other sensitive electronic devices. This is an especially important issue in hospital environment where it can have serious consequences to patient care. Many detailed studies have been published on the electromagnetic interference (EMI) between cell phones and various medical devices (Klein and Djaiani 2003, Lawrentschuk and Bolton 2004, Morrisey 2004, Shaw et al 2004). Evidence suggests that cell phones can cause malfunction of medical devices, but only when used in close proximity. For critical life-support equipments, the recommended immunity level to electromagnetic interference is the IEEE CE 60601 standard which specifies 3 V/m in the cell-phone frequency range (AAMI 1997, Morrisey et al 2002). This limit can be exceeded by some cell phones operating at full power even at distances of more than 1 m away (Hietanen et al 2000). For cancer patients treated with radioactive implants in the United States, the U.S. Nuclear Regulatory Commission requires the use of personal dosimeter for hospital workers attending the patient. Direct reading of pocket dosimeters or electronic personal dosimeters must be recorded at the beginning and end of each shift (NRC 34.47). Recently we have received alarming reports from nurses that attend to radioactive implant patients about very high readings reported by their pocket dosimeter. Immediate surveys with Geiger counter revealed no stray seeds or other unusual results in the patient’s room. Furthermore, the pocket dosimeters were later found to be working as expected in monitoring radiation under controlled test condition. While investigating the cause of this puzzling behavior of the dosimeter, it was revealed that the attending nurse was engaged in cell-phone conversation while wearing the pocket dosimeter. Our subsequent investigation points to this EMI from the cell phone as the most likely source of this erroneous behavior of the pocket dosimeter. This technical


3

Figure 1. Cell-phone interference with pocket dosimeters. Cell phones can produce erroneous, and alarmingly large, ionizing-radiation reading in pocket dosimeters as demonstrated in this figure. This behavior is more pronounced when the antenna or other transmitter parts of the cell phone (typically located at the back of the unit) are close to the center part of the dosimeter (circled as sensitive area in the figure). The interference can generally be suppressed by increasing the distance between the cell phone and the dosimeter to be greater than about 30 cm.

note presents a systematic study of this interference. 2. Methods and Materials We have tested the EMI of 4 different pocket dosimeters of model Aloka PDM-203 (Perspective Scientific, www.perspective.co.uk) with cell phones of several models. Kawano and Ebihara (1993) have performed tests using a 60 Co source on a similar model from the same manufacturer (Aloka PDM-101). They have concluded that the directional dependence of the dosimeter response to ionizing radiation is minimal. It is also mentioned in their paper that the PDM-101 dosimeter is nickel plated to shield it from external electromagnetic waves. The cell phones that we have used in this study are of the following models: palmOne Treo 600, Sanyo SCP-4900, and Samsung VI660. These phones are personalcommunication-systems (PCS) phones and some of them are also able to send and receive data (pictures, web contents) in addition to voice signals. Only interference during voice communication mode is tested in this study, however. The phones are digital dual-band devices and operate in both the 800-MHz and 1900-MHz CDMA (code division multiple access) frequency bands that are used in the United States. In CDMA technology, the voice signals are digitized and sent in small pieces over a number of discrete frequencies


4

Figure 2. Setup for measurement of EMI between the cell phone and the pocket dosimeter. The phone is placed at the edge of a styrofoam box with its antenna portion protruding out of the edge. The pocket dosimeter is placed below it flush to the edge of the styrofoam box. For field-intensity measurement, we placed a fieldintensity detector (also shown in this figure for display purpose—the field detector is removed during EMI measurement) in place of the pocket dosimeter.

available for use at any time in the specified range. Multiple calls are therefore overlaid on each other on each frequency channel. The phones can be readily shown to be capable of producing erroneous readings in the pocket dosimeter. In figure 1 we show an example of this EMI in which the dosimeter reported a reading of 4227 mrem after being in close proximity of an operating cell phone for about two minutes. For more careful measurement we have followed the setup shown in figure 2. The phone is placed at the edge of a styrofoam block with its antenna portion jutting out of the edge of the block. The dosimeter is placed at the level of the block’s base and its display is reset to zero after each measurement. The distance between the two devices is measured from the center of the antenna to the center of the dosimeter. The dosimeter-reading data are taken by dialing a public information line for 1 minute. Both the dosimeter and the cell phone are left untouched after the dial button is pressed so there is no change of distance or orientation between the two devices during the measurement. The data have been measured at the office area of our radiation oncology department which is located at the ground level of a hospital. The signal reception is relatively good, although not optimal, and no break up in transmission was observed during the measurement. Note that patients with radioactive implants are usually placed in a hospital room with the thickest walls to minimize the leakage radiation to their surroundings. Our measurement setup at a

5


Table 1. Typical readings of the pocket dosimeter in the vicinity of an operating cell phone. Three readings are shown for each setup and each measurement was taken for 1 minute call duration. The distance between the dosimeter and the cell phone for each setup is shown in the leftmost column. We have performed measurement both with the display side of the dosimeter (FRONT) and its pocket-clip side (BACK) facing up toward the cell phone. For the Treo-600 phone, no EMI is detected for the 10 cm setup, as is also the case for the Sanyo phone with the SN-50218 dosimeter. Phone model (SETUP, distance)

Pocket dosimeter (3 readings for each setup) SN-50218 SN-50220 SN-50226 SN-50231

Treo 600 (FRONT, 1 cm)

141 688 507

203 165 899

534 261 104

785 681 1053

Treo 600 (BACK, 1 cm)

54 58 72

64 186 154

217 68 67

225 301 245

Sanyo SCP-4900 (FRONT, 1 cm)

1792 1799 463

1559 587 1749

4474 5259 6934

1176 2119 1870

Sanyo SCP-4900 (BACK, 1 cm)

747 486 863

5377 792 1322

1427 1674 831

1176 2119 1870

Sanyo SCP-4900 (FRONT, 10 cm)

0 0 0

50 55 63

62 50 36

87 27 35

location with less than optimal cell-phone signal is thus appropriate for emulating the clinical situation. In addition to the readings of the pocket dosimeter, we have also measured the intensity of the electromagnetic field emitted by the cell phone. For this measurement we have used an HF-Detektor II Profi detector (Aaronia AG, Germany, www.electrosmog.de). This field detector, which is shown in figure 2, is capable of measuring the electromagnetic wave in average or peak mode. The average mode has been used in this study. The measured field intensity is reported as one of several LEDs (light emitting diodes), each of which is associated with a particular dB value relative to a calibrated intensity. For our field-intensity measurement, we placed the detector in place of the pocket dosimeter in figure 2. The LEDs that lit up for the phones that we tested correspond to the following intensity levels: 0.05, 0.3, 2.4, and 19 (in units of mW/m2 ). Although this detector is not capable of reporting an intermediate level between these values, it was noticed that during some measurements the intensity rapidly vacillated between two adjacent levels. For such cases, we will report the intensity to be in the range defined by the corresponding two levels.

6


Table 2. Typical field-intensity readings in the vicinity of an operating cell phone measured with HF-Detektor II Profi detector. Phone model Treo 600 Treo 600 Sanyo SCP-4900 Sanyo SCP-4900 Samsung VI660 Samsung VI660

Distance (cm)

INITIAL (mW/m2 )

NORMAL (mW/m2 )

1 10 1 10 1 10

2.4−19 0.3−2.4 19 0.3−2.4 19 2.4

0.3−2.4 0.05−0.3 2.4−19 0.05−0.3 2.4−19 0.3−2.4

3. Results Typical readings of the pocket dosimeter that we obtained in this study are shown in table 1. The first column in table 1 displays the phone model, setup, and phonedosimeter distance. Three readings are displayed for each setup. We have performed measurements with the digital-display side of the dosimeter facing up toward the cell phone (this setup is labeled as FRONT setup) and also with the pocket-clip side of the dosimeter facing up (labeled as BACK setup). It can be seen from the data that the front setup typically registers higher readings than the back setup. Furthermore, although the measurement setup is fixed, the data contain significant fluctuations. Since the measurement was performed by calling a public information line that simply plays a recorded message, the voice data transmitted to the cell phone is expected to be similar for each measurement. Thus the fluctuations are probably due to either the variation of voice data transmitted out by the cell phone or the concurrent level of network traffic during the call. It should be noted, however, that during the measurement each dosimeter registers a non-decreasing value as a function of calling time. Table 2 lists the electromagnetic field intensity at the position of the pocket dosimeter in the EMI measurement, measured using the Profi detector. Two values of field intensity are listed for each setup in table 2. The numbers in the column labeled INITIAL are the (higher) average intensity during roughly the first two seconds of the call when the phone is trying to establish a two-way channel with the base station. Those in the column labeled NORMAL correspond to the intensity during the rest of the call. Note that the average signal intensity in the initial phase is about an order of magnitude stronger than during the remainder of the call. This is particularly noticeable in the EMI measurement for the 10-cm setup shown in the last row of table 1. In this measurement all of the reading registered by the dosimeter was accumulated during the initial 2-3 seconds of the call with no subsequent increase in reading.


7

4. Discussions We have shown that cell phones can introduce significant EMI with pocket dosimeters. The actual level of interference can vary significantly among different models and it also varies with time during each call. The EMI is also dependent on the actual location of the cell phone relative to its nearest base station. Measurements of EMI conducted with the Treo-600 phone at an open field with very good reception signal produced zero to very small reading (typically less than 5 mrem/minute). Lacking detailed information of its design, the specific components of the pocket dosimeter that are sensitive to the EMI can only be speculated. The fact that the reading is non-decreasing with time suggests that the electronics of the pocket dosimeter treat the microwave signals from the cell phone similarly to to the input signals from its semiconductor detector component. It may also be of relevance to point out that the cell-phone signals involve fast digital modulation of the carrier microwaves. EMI tests between the pocket dosimeter and the leakage radiation from typical microwave oven (which operates with 2.45 GHz frequency, close to the 1.9 GHz frequency of CDMA cell phones, but in continuous mode) of comparable intensity typically gives much less reading than those shown in table 1. For shielded coaxial cables of nuclear detectors, it is known that undesirable transient signals can be induced if nearby equipment involves fast switching of large currents (Knoll 1989). Although significant EMI to the dosimeter can be caused by cell phones, simple measures based on radiation-protection principles (distance, time, and shielding) can be taken to mitigate it. Increasing the distance between the operating cell phone and the dosimeter is an effective way of resolving the interference. We have found that distances greater than about 30 cm are typically sufficient to achieve this suppression for the cell phones that we tested. The EMI can also be prevented by enclosing the dosimeter in a conductive shield. Placing it in an anti-static bag typically used for computer parts, e.g., can completely suppress the EMI while maintaining the detector’s sensitivity to ionizing radiation. This shielding should ideally be incorporated into the design of the pocket dosimeter by the manufacturer. Acknowledgements—We thank Susan Payne for useful reference and library assistance. References American Association of Medical Instrumentation 1997 Technical information report No. 18, Arlington, VA Hietanen M, Shibakov V, and Hallfors S 2000 Safe use of mobile phones in hospitals Health Phys. 79 (Suppl. 2) S77–S84 Industry Standard 2000 Look Ma, No Wires 1995–2005 Industry Standard September 2000, 205 Kawano T and Ebihara H 1993 Performance test of PDM-101 electronic pocket dosimeters using a 60 Co source Health Phys. 313–317 Klein A A and Djaiani G N 2003 Mobile phones in the hospital–past, present and future Anaesthesia 58 353–357


8

Knoll G F 1989 Radiation Detection and Measurement 2nd Edition John Wiley and Sons, New York, p. 559 Lawrentschuk N and Bolton D M 2004 Mobile phone interference with medical equipment and its clinical relevance: a systematic review Med. J. Aust. 181, 145–149 Morrissey J J, Swicord M, and Balzano Q 2002 Characterization of electromagnetic interference of medical devices in the hospital due to cell phones Health Phys. 82, 45–51 Morrissey J J 2004 Mobile phones in the hospital: Improved mobile communication and mitigation of EMI concerns can lead to an overall benefit to healthcare Health Phys. 87, 82–88 Shaw C I, Kacmarek R M, Hampton R L, Riggi V, Masry A E, Cooper J B, and Hurford W E 2004 Cellular phone interference with the operation of mechanical ventilators Crit. Care Med. 32, 928–931 U.S. Census Bureau 2003 Statistical Abstract of the United States p. 732 U.S. Nuclear Regulatory Commission 10 CFR.34 §34.47 Licenses for industrial radiography and radiation safety requirements for industrial radiographic operations http://www.nrc.gov/readingrm/doc-collections/cfr/part034/part034-0047.html