Addressing the Risk of EMC Problems with Mobile Radio Transmitters

A key element of prudent risk management is the recognition that safe operating distances may not be adequately covered by current immunity standards.

As usage of mobile telecommunication systems continues to increase, the risk of electromagnetic compatibility (EMC) problems associated with the use of mobile radio handsets grows, particularly in sensitive areas such as hospitals. The nature and extent of the risk, the adequacy of current immunity standards for addressing it, and means of managing the risk are examined in this article.

Mobile Telecommunication Systems

Types of mobile telecommunication systems in use today range from low-power two-way radio pagers to high-power analog private mobile radio handsets. Modern digital systems such as DECT (digital European cordless telecommunications), GSM (global system for mobile communication), and TETRA (TErrestrial Trunked RAdio) employ a cellular structure in which a base station is used to provide radio access over a relatively small geographic area. This allows use of a single bandwidth in multiple cells, thereby increasing the capacity of the overall system.

A number of techniques allow multiple users to access the system simultaneously (see Figure 1). In frequency division multiple-access (FDMA) systems, each user is assigned a different frequency channel on which to operate. In time division multiple access (TDMA), time is divided into defined periods, and different users use different time slots. TDMA systems therefore transmit bursts of power with a fixed pulse-repetition frequency known as the frame rate. GSM and TETRA are actually hybrid systems in which both FDMA and TDMA are employed.

Figure 1. A schematic comparison of FDMA, TDMA, and CDMA multiple-access techniques.

Third-generation systems will use a spread-spectrum technique known as code division multiple access (CDMA) to achieve both multiple-user access and flexible data rates. Several users are allocated the same wide-bandwidth channel; however, their signals are spread by means of a special code that enables them to be separated in a receiver that uses decorrelation techniques.

The planned European third-generation system—universal for mobile telecommunications system (UTRA)—defines a terrestrial radio access network that combines CDMA with time and frequency division duplex modes. The frequency division duplex (FDD) mode allocates the downlink and uplink in two separate frequency bands. This provides a symmetrical data channel with long spreading codes for global access to low- and medium-data-rate services. A separate time division duplex (TDD) mode allocates the uplink and downlink in the same frequency band but with transmission divided into time slots in a way similar to TDMA. This allows both multiple access by different users in different time slots and asymmetrical channels in which the downlink and uplink have different data rates. The TDD mode in UTRA is intended to be used for short-range, high-data-rate services such as Internet access.

Table I summarizes the important properties of a half-dozen key mobile radio systems. Their frequency bands, powers, and modulation schemes are defined by telecommunications standards. As can be seen from the table, a large range of parameters must be considered when the potential for EMC problems is being assessed.

The General EMC Problem

An EMC problem can arise when a mobile telecommunication transmitter is used in the vicinity of other electronic equipment. A piece of equipment such as a PC can become an unintentional radio receiver interacting with an intentional radio transmitter, or possibly several radio transmitters. The electromagnetic fields from mobile transmitters can be quite complex and, by comparison with unintentional emissions from nonradio devices, strong. The essence of the problem is twofold: (1) The radio transmitter is mobile and therefore represents a time-varying statistical threat. One cannot know absolutely where or when a mobile handset will be used. And (2), a mobile handset can be brought into very close proximity with a piece of equipment. The intentional radiation from the mobile instrument will then dominate the threat environment of the equipment.

Cellular radio base stations, although they operate at much higher powers than the mobile handsets, represent a fixed threat and are not likely to be in very close proximity to general electronic equipment.

What circumstances and conditions determine whether there is a risk of interference in such a situation? The LINK Personal Communications Programme of the U.K. Department of Trade and Industry has identified the following parameters as being most important:¹

• The transmitter power, which determines the overall strength of the electromagnetic field.
• The range to the victim equipment, which determines the strength of the electromagnetic field incident on the affected equipment.
• The number of transmitters; if more than one is present, their number and positions jointly define a complex threat environment.
• The frequency band or bands in which the transmitter operates, which contributes to a determination of how strongly the fields couple to a piece of equipment.
• The data modulation scheme, which can have an effect on EMC, although it is generally not an important parameter.
• The multiple-access scheme used for multiuser access by the radio system; for example, TDMA, FDMA, or CDMA, each of which has distinctive types of signal characteristics that may have an impact on EMC.
• The type of power control, which many mobile radio systems employ to minimize power consumption. Controls include discontinuous-transmission (DTX) protocols, which switch the transmitter off when no data are available for transmission, and which can generate low-frequency characteristics in the spectrum of the radio signals.
• The immunity profile of the equipment, which is very important but difficult to predict.

To assess the risk of interference between a mobile telecommunication system and other electronic equipment, each of these parameters must be considered.

Powers and Field Strengths

The power class of the transmitter generally determines the handset power. For example, TETRA defines both 1-W and 3-W handheld terminals, whereas for GSM900, the only power class used in practice for handheld devices defines a peak power of 2 W. The powers are generally higher for vehicle-mounted transmitters, and base stations can transmit many tens of watts. The Dolphin TETRA network in the United Kingdom, for example, will be using 25-W base stations. The peak and average powers for the mobile handsets in several radio systems are presented in Table II.

System	Power Class	Maximum Peak Power (W)	Maximum Average Power (W)
TETRA	3	3	0.75
TETRA	4	1	0.25
GSM900	4	2	0.25
GSM1800	1	1	0.125
UTRA/TDD	3	0.125	0.08
UTRA/FDD	4	0.125	0.125
Table II. Peak and average powers of the handset in various mobile radio systems.

To compare the electric-field strength generated by a given class of mobile terminals to the immunity levels in current standards involves the use of a simple source model that relates the root-mean-square (rms) electric-field strength, E, to the peak total radiated power, P (actually the effective isotropic radiated power), and the range from the transmitter, d, according to

Figure 2 shows how the electric-field strength varies as a function of the range from the transmitter for various transmitter powers. The graph includes the 3- and 10-V/m immunity standards as baselines. As can be seen, the field strength for all classes of mobile transmitters exceeds the immunity standards at short distances.

Figure 2. Variation in electric-field strength with distance from the transmitter for different transmitter powers.

For a 1-W mobile device, at distances closer than 2 m the field strength exceeds 3 V/m, and at around 60 cm the field reaches the 10-V/m standard. For a 10-W transmitter the corresponding distances are 5.5 and 1.8 m, respectively. Note that this simple model of the source is valid only at a distance greater than a wavelength, which is 0.7 m for TETRA and 0.3 m for GSM900. The immunity standards therefore do not exclude the possibility of conforming equipment experiencing interference from close-range mobile transmitters.

The model can be used to define a safe operating distance, d_safe, from a mobile transmitter with power P for a piece of equipment with an immunity level of E_immune by employing the equation

This is illustrated in Figure 3, where the safe distances for both 3- and 10-V/m immunity standards are shown. However, this should be considered only as a broad guide, not as a strict definition that guarantees no interference will occur. As outlined above and discussed in detail below, the power from a mobile handset is not the only factor that determines EMC.

Figure 3. The safe distance for operation of mobile radio handsets as a function of the transmitted power.

Base stations operate at much higher power than handheld transmitters. Figure 2 also shows the field strength of a 25-W transmitter. The maximum field is around 3 V/m at a range of 10 m, which is the same as a 1-W mobile transmitter at 2 m. However, given the sites at which base stations are typically installed, it is much more likely that the equipment of concern will be less than 2 m from a mobile handset than less than 10 m from a base station. Base stations are also fixed. Problems can be dealt with via installation guidelines. This may become more of an issue with third-generation systems that will make more extensive use of picocells that utilize a large number of low-power base stations in areas requiring high capacity.

Power Variations

Immunity standards consider carrier-wave and simple amplitude-modulated threats. The signals emanating from one or several mobile transmitters are much more complex. They contain a range of frequencies that are generated by various properties of the telecommunication system. Table III gives the characteristic frequencies of several system parameters, including the beat-like modulation that occurs when multiple transmitters are used in the vicinity of a piece of equipment. An EMC problem usually arises when one of these characteristic frequencies coincides with the operating frequency of the equipment.

Property	TETRA	GSM900	GSM1800	UTRA FDD	UTRA TDD
TDMA (Hz)	17	217	217	N/A	100
Power control (Hz)	17	217	217	1600	800
Data modulation (kHz)	25	200	200	3840	3840
Multiple phone (MHz)	0.1–10	0.4–25	0.4–40	5–200	5–200
Carrier (MHz)	400	900	1800	2000	2000
Table III. Characteristic frequencies of various system properties.

The characteristic frequencies of the system that lead to amplitude modulation of the signal are particularly important because these are more likely to cause EMC problems than phase-modulation effects.

The frequency of the TDMA bursts differs significantly between TETRA and GSM. The burst rate of 17 Hz in TETRA is lower than that of GSM by a factor of 10. This low-frequency TDMA structure lends itself particularly to causing EMC problems. The classic example of this is that demodulation of the 217-Hz burst rate from GSM phones by some hearing aids causes a buzzing sound.^2,3 The burst rate of 17 Hz in TETRA is probably low enough to prevent this particular problem, though it may cause other effects in equipment with very-low-frequency passbands.

EMC Standards for Immunity

Although equipment immunity is difficult to determine, all equipment must satisfy applicable EMC standards in order to be placed on the market. This is the only generic way to assess the immunity of equipment currently in use.

In Europe, most equipment must comply with the EMC Directive, which references the generic standards EN 50082-1 and EN 50082-2 for radiated immunity. Medical equipment is governed by a separate standardization process defined by the Medical Devices Directive and Active Implanted Medical Devices Directive, which reference IEC 60601-1-2 for radiated immunity.⁴ Table IV summarizes the required immunity levels for equipment under these standards.

Standard	Frequency (MHz)	Level (V/m)	Notes
EN 50082-1:1992 Generic standard: Residential and light industrial	27–500	3	80% AM at 1 kHz
EN 50082-1:1997 Generic standard: Residential and light industrial	80–1000	3	80% AM at 1 kHz
	900	3	Keyed carrier, 50% duty cycle, repetition rate 200 Hz
EN 50082-2:1995 Generic standard: Industrial	80–1000	10	80% AM at 1 kHz
IEC 60601-1-2:1993 Collateral standard: Medical equipment	26–1000	3	80% AM at 1 kHz
Table IV. Summary of radiated immunity standards.

In broad terms, the current standards specify that equipment for heavy industrial environments should be immune to 10 V/m rms, and other classes of equipment to 3 V/m. The frequency range for the standards is generally 30 or 80 MHz to 1 GHz, which covers the TETRA and GSM900 bands. It is interesting to note that GSM1800, UTRA, and other mobile telecommunication systems operating above 1 GHz are beyond the range of the current generic immunity standards.

The standards generally allow for carrier-wave threats and simple modulated threats (80% AM in signal-processing passbands of the equipment being tested). The 1997 update of EN 50082-1 also references ENV 50204, which defines a test for keyed carrier signals with parameters established to test immunity to GSM900 mobile phone signals.⁵ ENV 50204 also makes provision for tests of immunity against DECT cordless phone signals, but these are not used in the generic standard because the frequency is above 1 GHz. This revised standard does not cover lower TDMA frame rates such as that used by TETRA, nor complex signals generated by multiple radio sources. The immunity limit of 3 V/m corresponds to a safe distance of about 2.6 m for a GSM900 handset.

The Medical Devices Directive (MDD) recognizes that for certain types of equipment, particularly sensitive diagnostic equipment, it is very difficult to prevent interference from radio-frequency sources without affecting normal equipment operation. The MDD allows a product to fail on the condition that clinical benefit can be demonstrated and that adequate documentation and operating guidelines are provided.

The problems of mobile transmitters are being addressed by the new draft standard IEC 60601-1-2 for medical equipment.⁶ The frequency range of the tests has been extended to 2 GHz, with a 10-V/m immunity level from 800 MHz to 2 GHz for life-support equipment. The immunity level remains 3 V/m for 80–800 MHz for life-support equipment and over the entire range for non-life-support equipment. The 800-MHz changeover point in immunity levels is interesting. It ensures that GSM900 is included in the higher immunity level range but fails to recognize TETRA at 400 MHz.

The new standard will also require manufacturers to fill out pro forma tables to provide more information to equipment operators on the EMC characteristics of their equipment, including recommended safe operating distances from intentional radiators. Clearly, this is an important improvement in the EMC standardization process. It will make available information necessary for assessing risk and setting policies and for which current standards make no provision.

The Hospital Environment

The hospital is an interesting specific environment to consider because it is a safety-critical setting containing a wide variety of electronic equipment. Medical equipment falls broadly into two categories: therapeutic and diagnostic. Therapeutic equipment is used for providing some form of treatment, such as infusion pumps that administer a fixed quantity of drug at a specific rate. Diagnostic equipment is used to monitor biological functions. Equipment such as an electrocardiograph requires sensitive transducers to measure physiological signals with bandwidths of about 100 Hz and amplitudes less than 1 mV. In addition, hospitals contain a variety of general information-technology equipment that may have an indirect effect on patient care. Any equipment with a medical function must meet the requirements of the MDD.

Potential interference problems from mobile handsets in hospitals first came to attention in the United Kingdom in 1994 when the Medical Devices Agency (MDA) issued a safety advice notice.⁷ This led to a number of restrictions on mobile phone use in hospitals and prompted a detailed statistical study, whose results were published in 1997, to quantify the problem.⁸

The study looked at 178 devices covering the whole spectrum of medical equipment. Each device was subjected to the electric field from 12 types of radio handsets, including cordless phones, cellular phones, and emergency service handsets. The power level of the handsets was uncontrolled; however, the equipment was used in real environments and therefore provided a good indication of the extent of possible problems. Ninety percent of the equipment was apparently unaffected at a range of 1 m. At 0.5 m, 6.6% of tested equipment showed effects that could have affected patient care.

An analysis of these effects in terms of the type of handset proves interesting. The MDA constructed a statistical model of the data to predict the distance at which different classes of equipment would have a 5% chance of experiencing interference. Results appear in Figure 4. It is clear from the trend that the parameter dominant in determining the average EMC effect is the power. High-power analog systems produce the greatest effects; cellular mobile phones create fewer problems; and cordless phones generate problems only at effectively zero range, if at all.

Figure 4. Results of the MDA study of medical equipment susceptibility to interference from mobile radio handsets.

This analysis would suggest that 1-W TETRA mobile instruments will be similar to GSM900 devices, or even slightly better, in terms of their average EMC effects on other equipment. Of course, this is a general statistical argument that overlooks the cases where equipment has a specific susceptibility to some characteristic of the field emanating from a particular mobile system.

The LINK Personal Communications Programme has undertaken some rigorous immunity tests of medical equipment using simulated TETRA signals, as noted in Appendix I of the report cited for this article. One of the pieces of equipment tested showed a 10-fold increase in susceptibility between an unmodulated 400-MHz carrier and the same carrier with 17-Hz pulse modulation. This corresponds to a 10-fold increase in the required safe operating distance. It is interesting that the equipment in question was also immune to simulated GSM900 signals similar to those defined by ENV 50204 and to a signal using the GSM900 frame rate of 217 Hz at the TETRA carrier frequency of 400 MHz. In this case, the equipment was specifically susceptible to the very low (17-Hz) frame rate of the TETRA signal. This highlights the potential pitfalls of basing a safe operating distance on immunity levels that are not derived from representative threat fields.

Mitigation

What can be done to alleviate the risk of interference from mobile transmitters in sensitive areas? The following actions are recommended.

• Increase awareness of potential problems.
• Manage mobile handset usage in critical areas.
• Define safe operating distances. (Although the problems with this approach have been discussed, an exclusion zone of around 2 m will provide protection against the majority of possible effects.)
• Provide guidelines for installation and operation for sensitive equipment that may be used in the vicinity of mobile transmitters. (The MDA advice for ambulances is an example.⁹)
• Test safety-critical equipment against the intended operating environment.
• Improve the EMC design of electronic equipment.

Conclusion

Although no evidence of widespread problems caused by the use of mobile radio handsets exists, operative EMC standards leave open windows of opportunity for EMC problems to occur. Many current and future mobile radio systems, such as GSM1800 and UTRA, lie outside the frequency range of the immunity standards now in place. These standards also employ simple threat fields that do not include the complex modulations present in today's digital telecommunication systems. In particular, low-frequency variations of the signal envelope from TDMA, DTX, and other system features are known to cause interference in some types of equipment. Most of these problems occur only when the transmitter is very close to the equipment. An exclusion zone of 2 m around sensitive equipment will therefore provide protection against most potential effects.

Acknowledgment

Work done for this article was funded by the Engineering and Physical Sciences Research Council as part of the U.K. Department of Trade and Industry LINK Personal Communications Programme.

References

1. LINK Personal Communications Programme, U.K. Department of Trade and Industry, Electromagnetic Compatibility Aspects of Future Radio-Based Mobile Telecommunication Systems, final report (DTI: London; York, U.K.: LINK Personal Communications Programme, 1999). Also available on the Internet: http://www.emc.york.ac.uk/reports/linkpcp/index.html.

2. GF Pedersen, "Amplitude Modulated RF Fields Streaming from a GSM/DCS-1800 Phone," Wireless Networks no. 3 (1997): 489–498.

3. F Han and J Nuutinen, "Analysis of Spurious Spectrum Due to RF Bursting Signals in TDMA-Based Wireless Communication Systems," in Proceedings of the IEEE International Symposium on Electromagnetic Compatibility (Denver: 1998), 393–398.

4. IEC 60601-1-2, Section 36.202, International Electrotechnical Commission, Brussels, 1993.

5. ENV 50204, "Radiated Electromagnetic Field from Digital Radio Telephones—Immunity Test," CENELEC (European Committee for Electrotechnical Standardization), Brussels, 1995.

6. IEC 60601-1-2/62A/206/CD, draft 2nd ed., International Electrotechnical Commission, Brussels.

7. MDA Safety Action Bulletin SAB (94)49, "Portable, Cordless, and Cellular Telephones: Interference with Medical Devices," Medical Devices Agency, London, November 1994.

8. MDA Device Bulletin DB 9702, "Electromagnetic Compatibility of Medical Devices with Mobile Communications," Medical Devices Agency, London, March 1997.

9. MDA Device Bulletin DB 1999(02), "Emergency Service Radios and Mobile Data Terminals: Compatibility Problems with Medical Devices," Medical Devices Agency, London, May 1999.

Ian Flintoft, PhD, is a research fellow in Applied Electromagnetics at the University of York (U.K.). He is currently active in a number of areas of research including computational electromagnetics, immunity of digital systems, EMC in complex and distributed systems, and EMC aspects of telecommunication systems. He holds a PhD in physics from the University of Manchester. He can be reached at idf1@ohm.york.ac.uk. Martin Robinson, PhD, is a lecturer at the university. His research includes EMC design, dielectric measurements, and the interaction of electromagnetic radiation with biological tissues. He holds a PhD in dielectric imaging from the University of Bristol. He can be reached at mpr@ohm.york.ac.uk. Stuart Porter, PhD, is also a lecturer at the university. His research interests include computational electromagnetics, particularly as applied to radio-frequency and microwave problems; antenna design; computational and computer-aided tools for EMC design; application of evolutionary computation to antenna design and EMC; computational acoustics. He holds a DPhil from the University of York in theoretical physics. He can be reached at sjp1@ohm.york.ac.uk. Andrew Marvin is professor of applied electromagnetics at the university and technical director of York EMC Services Ltd. He is a member of both IEE and IEEE and is a chartered engineer in the U.K. His research interests include EMC measurement techniques, electromagnetic shielding measurements, EMC antenna design and equipment design techniques for EMC. He is currently chairman of the management committee of COST Action 261 (EMC in Complex and Distributed Systems) and contributes to various national and international standards committees. He can be reached at acm@ohm.york.ac.uk.

Back to September/October Table of Contents