==Phrack Magazine== Volume Four, Issue Forty-Four, File 10 of 27 **************************************************************************** Protective Measures Against Compromising Electro Magnetic Radiation Emitted by Video Display Terminals by Professor Erhart Moller University of Aachen, Aachen, Germany 0. Introduction Compromising electromagnetic radiation emitted by machinery or instruments used in data processing or communication engineering can be received, decoded and recorded even across large distances. It is also possible to recognize the data or information which was processed and transmitted by the emitting instrument as text in clear. Compromising emitted electromagnetic radiation thus jeopardizes the protection and security of data. The Laboratory for Communication Engineering at the Fachhochschule Aachen is developing protective measures against compromising emission of radiation. However, these protective measures can only be effective if they are derived from the characteristics, the effects, and risks of compromising emitted electromagnetic radiation. Therefore we first consider only the forms of appearance and the characteristics of compromising emitted electromagnetic radiation. 1. Compromising Emitted Electromagnetic Radiation In this context one often refers only to the so-called computer radiation. But this is only one form of compromising emitted electromagnetic radiation. There are three types of such emissions. 1.1. Types of Compromising Emitted Electromagnetic Radiation Figure 1.1 shows an n example of an arbitrary electric device with various electric connections: a power supply line, a high frequency coaxial transmission line, and a coolant line with in- and outflux. This device emits three types of compromising electromagnetic radiation: 1. electromagnetic radiation in form of electric and magnetic fields and electromagnetic waves; 2. electromagnetic waves on the outer surface of all coaxial metallic connections (shell waves); 3. electric interference currents and interference voltages in power lines connected to the device. Each of the three types can be transformed into the other two. For instance, shell waves can be emitted as fields or waves. On the other hand, electromagnetic waves can be caught by a nearby conductor and can propagate on it as shell waves. These phenomena are the reason for the difficult control of compromising electromagnetic radiation, and they imply that one must deal with all and not just one form of compromising electromagnetic radiation. Also, electromagnetic protection against compromising emitted radiation must deal with all forms of it. 1.2. Examples of Compromising Emitted Electromagnetic Radiation To exemplify the three types of compromising electromagnetic radiation we consider the monitor depicted in figure 1.2. 1.2.1. Compromising Electromagnetic Radiation Figure 1.3. shows the experimental set-up. The video display terminal is connected via the power line to the power supply. The power line is surrounded by absorbers so that the terminal can only emit electromagnetic radiation. The absorbers prevent the generation of shell waves on the power line. The dipole antenna of the television receiver is 10 m from the video terminal. Figure 1.4. shows the screen of the television receiver after it received and decoded the signal. Not only is the large FH=AC well readable but also the smaller letters. This demonstration yields the following results: * The video display terminal emits electromagnetic radiation; * Despite being within (standards committee) norms the emitted electromagnetic radiation can be received and decoded across a certain distance; * The electromagnetic radiation emitted by the terminal can be decoded into readable information and symbols on a television screen. Therefore, this emitted radiation is compromising. 1.2.2. Compromising Surface or Shell Waves The video display terminal and the television receiver are positioned as in figure 1.5. The power line of the terminal is surrounded by a current transformer clamp which absorbs the shell waves. The television screen shows again the picture seen in figure 1.4. The quality of the picture is often better than in the previous case. Another experiment would demonstrate that secondary shell waves can form on a nearby conductor. The emitted radiation is then caught by nearby conductors and continues to propagate as shell waves. These emissions also give good receptions but are almost uncontrollable along their path of propagation. 1.2.3. Demonstration of Compromising Emitted Radiation Through the Power Line Figure 1.6 shows the experimental set-up for the proof of compromising power supply voltages. The video display terminal acts as a generator whose current and voltage is entered into the power supply. Using a capacitive line probe, the entered signal can be retrieved and fed into the television receiver. This form of transmission is the known basis for intercom systems or so-called babysitter monitors where the signals are transmitted from room to room via the energy supply lines in a home. As in the case of electromagnetic radiation or shell waves, one obtains the same picture quality as in figure 1.4. 2. Facts About Compromising Emitted Radiation Protective measures against compromising emitted radiation are not only determined by the above-mentions\ed three types of compromising emissions but also by taking into account the following data: # level of intensity and spectral distribution; # frequency (emission frequency) and frequency range; # directional characteristics of the radiation. These data can then be used to derive the damping and the amplitude-frequency response for the protective measure and its location. 2.1. Emission Spectrum and Level of Intensity The spectral distribution of compromising emitted radiation depends on the frequencies used to generate the picture on a screen. The regular repetition of dots and lines gives rise to the video and line frequency which is found in the spectrum. However, the emission of video or line frequencies is not compromising since their knowledge does not yet give access to processed data. If the lines are covered regularly by symbols, a symbol frequency is obtained which is also detectable in the spectrum. A single symbol consists of a dot or pixel matrix. The dot matrix of the symbol @ is also known in figure 2.1 The electron beam scans the individual dots or pixels line-by-line and keys them bright or dark. This keying is done using the so-called dot or pixel frequency. For instance, the highest keying frequency is obtained by scanning the center of the @ symbol since there one has a long sequence of successive bright and dark pixels. It also follows from figure 2.1 that the keying is slower, i.e., the keying frequency is lower, along the upper part of the @ symbol because of a long sequence of only dark or bright pixels. It follows that the emissions due to the keying frequency are highly compromising since they give direct information about the structure of the picture. Until recently, the frequencies in the following table were used: video frequency 45 Hz - 55 Hz line frequency 10 kHz - 20 kHz symbol frequency 2 MHz - 5 MHz dot or pixel frequency 15 MHz - 20 MHz. The pulses for the electron beam are formed in the video part, i.e., the video amplifier, of the monitor. Therefore, the cathode-grid of the picture tube and the video amplifier are the main emitters of radiation. The upper diagram in figure 2.2 shows the calculated spectrum for the cathode-keying. It represents a sequence of dots from the center of the @ symbol using a dot-sequential frequency of 18 MHz. The diagram in the center of figure 2.2 shows the measured spectrum at the keyed cathode of the picture tube. The agreement between the calculated and measured spectrum for the frequency is clearly visible. However, the calculated and measured spectral representation differ in the form of the envelopes. In the measured spectrum one finds an amplitude increase between 175 MHz and 225 MHz. This increase is usually found in the same or similar form in monitors. The reasons for this amplitude increase are design, construction parts, and dimensions of the video display terminal. In the lower part of figure 2.2 we see the compromising radiation emitted by the terminal as measured at a distance of 10 m. The spectrum of the radiation emitted by the terminal is superimposed by broadcast, radio and interference spectra since the measurement took place on open ground. Despite this interference one can recognize the typical form of the cathode spectrum. The increase in the amplitude between 175 MHz and 225 MHz presents a particular risk since the television transmitters for Band III operate within this frequency range and all television sets are tuned to it (see figure 2.2). A comparison of the intensity level of the television transmitter with the level of the compromising radiation in figure 2.2 shows their agreement. It is therefore not very difficult to receive the compromising radiation in proximity of the emitter using only a regular television set with normal sensitivity. Figure 2.3 shows the spectral distribution of compromising shell waves emitted by the video display terminal. Here again one recognizes the particular form of the dot or pixel frequency. The height of the shell wave spectrum is much lower at higher frequencies than the height of the radiation spectrum. The shell waves have lower intensity in the range of broadcast television but higher intensity in the range of cable television. To receive the shell waves a television set must be cable-ready. Figure 2.4 shows the spectrum for the third type of emission: the compromising currents and voltages entering the power supply lines. It is very similar to the shell wave spectrum. The height of this spectrum at higher frequencies is even smaller than the shell wave spectrum. In order to receive any signal a cable-ready television set must be used. The intensity of the currents and voltages is so high that they can easily be received using a regular television set with normal sensitivity. 2.2. Frequency and Frequency Range It follows from figures 2.2, 2.3, and 2.4 that the best reception for the three types of emissions is for the following frequencies: compromising radiation approx. 200 MHz; compromising shell waves approx. 60 MHz; compromising voltages approx. 20 MHz. The video information of the picture on the monitor has a frequency range of half a spectral arc. The frequency range of the receiver must therefore be 10 MHz for all three types of emission. 2.3. Directional Characteristics of the Radiation Figure 2.5 shows the directional characteristics for compromising radiation emitted by a video display terminal inside a plastic casing. According to this diagram the lateral radiation dominates. The field intensity along the front and back direction is about 30% of the lateral intensity. The power of the emitted radiation along these directions is only about 10% of the power emitted laterally. The range for the emitted radiation along the front and back direction is therefore also reduced to 30%. This phenomenon suggests for the first time a protection against compromising radiation, namely proper positioning of the device. The compromising shell waves and power line voltages propagate according to the configuration of the lines. There is no preferred direction. 2.4. Range The range of compromising radiation emitted from a video display terminal is defined as the maximum distance between the emitting terminal and a television receiver and readable picture. The range can be very different for the three types of emitted radiation. It depends on the type of emitter and the path of propagation. The spectacular ranges for emitted ranges are often quoted - some of which do not always come from the technical literature - give in general no indication just under which conditions they were obtained. It is therefore meaningful to verify these spectacular ranges before using them. 2.4.1. The Range of Compromising Emitted Radiation The dependence of the field intensity on distance is illustrated in figure 2.6. The dependence of the range on the receiver used is shown at 25 m, 40 m, and 80 m. The field intensity at 25 m is just strong enough to receive a picture with an ordinary television receiver using the set-up in figure 1.3. If one uses a narrow-band television antenna or a noiseless antenna amplifier than the field intensities at 40 m and 80 m, respectively, are still strong enough to receive a legible picture. The flattening out of the curve at large distances suggests that the range can be increased to several hundred meters by using more sensitive antenna or better receivers. The range can also be increased through a high altitude connection, for instance, if both emitter and receiver are in or on a high rise. This was verified by an experiment involving two high rises separated by over 150 m. A very clear picture was received using a relatively simple antenna with G = 6 db. 2.4.2. Range of Compromising Shell Waves Measurements have shown that shell waves can propagate across a large area without any noticeable damping if only the surrounding metallic conductors extend also across the entire area. The propagation is reduced considerably by a metallic conductor that crosses metallic surfaces such as metal walls or metallic grids such as reinforcements in concrete walls. Dissipative building materials also damp shell waves. Lightweight construction such as the use of dry walls or plastic walls in large buildings increases the range of shell waves to about 100 m without the picture becoming illegible. 2.4.3 Range of Emissions Through Power Supply Lines In this case the conditions are even less clear than in the previous cases. It must be assumed that inside a building the compromising currents and voltages can be received through the phase of the power supply lines feeding the video display terminal . The possibility of receiving the signal through other phase lines by coupling across phases in the power supply line cannot be excluded. The range depends very much on the type of set-up and the instruments used. It is conceivable that a range of about 100 m can be obtained. 3. Protective Measures Protective measures fall into three categories: - modification of devices and instruments by changing procedures and circuitry; - heterodyning by noise or signals from external sources; - shielding, interlocking, and filtering. 3.1. Instrument Modification The instrument modifications consist of changing the signal processing method and the circuitry of the instrument. It is the objective of these measures to alter the spectral distribution and intensity of the emitted radiation in such a way that the reception by television sets or slightly modified television sets is no longer possible. For instance, a change of procedure could consist of a considerable increase in the dot or pixel frequency, the symbol and line frequencies. A reduction in the impulse amplitude and impulse slope also changes the reduction in the impulse slope also changes emission spectrum so that reception is rendered more difficult. However, the subsequent modification of the video display terminal has serious disadvantages of its own: First of all, the user of video display terminals does in general not possess the personal and apparative equipment to perform the modifications. To complicate things further, the so-modified instruments loose their manufacturer's warranty and also their permit of operation issued by governmental telecommunication offices. A subsequent instrument modification by the user is for these reasons in general out of question. 3.2. HETERODYNING STRATEGY We refer to a protective measure as a heterodyning strategy whenever the compromising emitted radiation is superimposed by electromagnetic noise of specific electromagnetic signals. The television set receives the compromising emitted radiation together with the superimposed noise of spurious signal. The noise or the spurious signal are such that a filtering out or decoding of the compromising emitted radiation by simple means is impossible. Since the noise and the spurious signal not only interfere with the television receiver of the listener but also with other television sets in the vicinity the heterodyning strategy is by all means in violation with the laws and regulations governing telecommunications. As far as is known, this is a protective measure only used under extremely important circumstances involving high government officials. 3.3 Shielding In contrast to the previously considered protective measures, shielding has two important advantages: * shielding protects not only against compromising emitted radiation but also against electromagnetic emissions which can enter data processing devices from the outside and cause interference; * furthermore, shielding neither violates the laws governing the use of telecommunications nor does it jeopardize the manufacturer's warranty. The term shielding is used here to describe, shielding, interlocking, and filtering. 3.3.1. Shielding Data The requirements on a shield are described by the shield damping. The shield damping is twenty times the logarithm of the ratio between the electric or magnetic field intensity inside the shield and outside the shield. Actual applications and individual situations may require different values for the shield. The shield data are derived from the so-called zone model. In the zone model one considers the type and intensity of the emitted radiation, the composition of the path of propagation, and the local accessibility for the receiver. The shield data not only influence the shield damping but also the frequency range of the shield's effectiveness. Figure 3.1 shows a diagram listing different types if shields according to regulations MIL STD 285 and 461B, NSA 656, and VG norms 95 375. 3.3.2. Applicability of Shielding Electromagnetic shielding can be used on emitting or interfered with instruments, on building and rooms, and on mobile cabins. 3.3.2.1. Shielding of Instruments The shielding of instruments though it can often be done very quickly and effortlessly is not without problems. In general but especially after subsequent installation, it can lead to a loss in design and styling of the shielded device. Openings in the shield, for instance for ventilation or control and operating elements, cannot always be sealed off completely. In this case they are emission openings with particularly high emission rates. Trying to maintain ergonometric conditions - good viewing conditions for the users - renders the shielding of screens especially difficult. If the casing of the instruments is not made of metal but of plastic, the following shielding materials are considered: metal foils, metal cloth, metal-coated plastics, electrolytical layers and coats of metallic paint or paste. Recently, the plastics industry is also offering metallized plies of fabric. Such glasses are for instance offered by VEGLA, Aachen. Ventilation openings are sealed off with metallic fabric of honey-comb wirings. Interlocking systems and filters on all leads coming out of the instrument prevent the emission of compromising shell waves and power supply voltages. 3.3.2.2. Building and Room Shielding There are some advantages in shielding buildings and rooms. The building and room shielding lies solely in the competence of the user. Minor restrictions dealing with the static of the building and local building regulations only occur with external shielding. Building and room shielding offers a protection that is independent of the instrument or its type. It is a lasting and effective protection. Maintenance is minimal, and subsequent costs hardly exist. Interior design and room lay-out are not changed. If one requires better shielding values or a building and room design which emphasizes better comfort than greater expenses and thus higher costs will occur. 3.3.2.3. Cabin Shielding Cabin shielding has all the advantages of building and room shielding. In addition, cabin shielding is not affected by the static of the building or local building regulations. Furthermore, cabin shielding requires less expenses and costs than building or room shielding. However, shielded cabins do not offer the same comfort or interior design as shielded buildings or rooms. 3.3.3. Shielding Components Electromagnetic shielding consists of three components: # the actual shield together with various structural elements as a protection against emitted radiation; # the interlocking of all non-electric and electric supply lines to protect against shell waves; # electric filters at all supply lines to protect against compromising power supply voltages. 3.3.3.1. The Electromagnetic Shield The shield consists of the hull and the shielding structural elements. 3.3.3.1.1. Shield Hull - Method and Construction In general, one uses metal sheets or metal foil to construct electromagnetic shields for buildings and rooms. If one lowers the requirements on the shield damping and the upper limit frequency then screen wire, metallic nets, and - if properly constructed - even the reinforced wire net in concrete can be used; the obvious disadvantage is that the settlements or movements of the building can cause cracks that will render the shield ineffective. Therefore, only metal shields or strong wire netting is used for the construction of electromagnetically shielded cabins. The building or room shield can be built using several construction principles. Figure 3.2 above shows the essential construction principles. For the Sandwich construction, the shield is between the outer and inner layer of the wall. A new type of construction uses the Principle of the Lost Form. The shield itself which consists of 3 to 5 mm thick sheet iron is used as an inner layer in the manufacturing of concrete walls. The sheets touch one another and have to be welded together at the contact points. If the building or room shields he\ave to satisfy a special purpose then they have to be grounded at only one point; they have to be assembled in such a way that they electrically insulate against the building or room walls. The so-called inner shields offer this protection. In simple cases, the inners shield is placed on top of the walls maintaining insulation by using a special underneath construction. However, this space-saving and simple construction has a disadvantage; the part of the shield that faces the wall such as corrosion, settling or moving of the building, or damages due to work on the exterior of the building can no longer be detected. The use of non-corrosive shield material or sufficient back ventilation of the shield protects against corrosion in these cases. The self-supporting inner shield is suspended from a supporting grid construction. This construction can be similar to a cabin construction. In the case of large rooms, such as halls, one should use a truss for statistical reasons. The self-supporting inner shield has the advantage of accessibility, although the usable room volume has been decreased. In rooms where the shield is exposed to only slight mechanical wear and tear and not required to shield completely, shielding metal foil is glued directly to the wall and welded at the contact points. The floor construction is almost the same for all four construction principles. It is important that the floor onto which the shield is placed is protected from humidity and is even. In the case of electrically insulating layers of, for instance, laminated paper or PVC are first put on the floor. The ceiling construction depends on the specific requirements and necessities. The ceiling shield can be a suspended metallic ceiling or a self-supporting ceiling construction. 3.3.3.1.2. Shield Construction Elements Construction elements which seal off viewing openings or access openings are called shield construction elements. Access openings are doors, gates, and hatches. Viewing openings are windows. The shielded doors, gates, and hatches serve two purposes: first to close off the room, and second to shield the room. The door, gate, or hatch shield is in general made of sheet iron. Passing from the door or gate shield to the room shield causes shield-technical problems. A construction which is due to the company of TRUBE & KINGS has proven to be especially effective for this kind of problem (see figure 3.3). The set-on-edge door shield, the so-called knife, is moved into a U-shape which contains spring contacts. The difference between this and other available constructions is that the knife is not moved into the spring upward. This construction reduces the wear and tear of the transition point between door and room shield and thus increases the durability of the construction which implies a better protection and higher reliability. This construction by TRUBE & KINGS satisfies the highest requirements on shield damping. Windows in shielded room are sealed off with the shielding glass or so-called honey-comb chimneys. It si understood that these windows are not to be opened. Figure 3.4 shows the cross-section of a glass especially developed by VEGLA for data processing rooms. The glass consists of multiple layers which are worked into a very fine metallic net and an evaporated metallic layer. The thickness of the wire is in the range of a few micrometers so that the net is hardly visible. This glass can also be manufactured so that it is rupture- and fire-resistant and bullet-proofed. Using glass one can reach shield dampings in the medium range (refer to figure 3.1). Specially manufactured glass reaches even higher shield dampings. Figure 3.4 also shows the so-called honey-comb chimneys as manufactured by SIEMENS. Visibility and the comfort of light are highly restricted. But the advantage is that this type of shielding satisfies the requirements for highest shield damping. 3.3.3.2. Interlocking All non-electric supply lines leaving a shielded room must be interlocked in order to protect against the propagation of shell and surface waves. Water pipes, heating pipes, pneumatic and hydraulic pipes are connected via rings to the metallic shield. Depending on the required frequency range, the pipe diameter is also subdivided by filter pieces. At high frequencies on can achieve dampings of up to 100dB using such interlocking devices. The ventilation of shielded rooms may cause problems. Problems will occur if shield dampings up to the highest frequencies are required. In this case one has to use two-step ventilation filters. The first step consist of adding concave conductor filters which work for the frequencies up to 200 GHz, the second step of adding absorber filters which protect against compromising emitted frequencies above 200 GHz. Figure 3.5 shows the set-up for the above-described ventilation lock which is due to the SCHORCH. 3.3.3.3. Electric Filters Filters must be put on electric power supply lines, telephone wires, and data processing supply lines at the room exit point. The filters have to be installed at the shield. The filters used here are the same as the ones shown in the area of electromagnetic compatibility. 4. Summary Electric devices used in data processing, data transmission and data handling emit electromagnetic radiation, electromagnetic shell and surface waves, and currents and voltages in power supply lines, telephone wires, and data supply lines. If this emitted radiation carries actual data or information from the data processing device then it is compromising. Using a television receiver, it is very easy to receive, decode and make these compromising emissions legibly. Several possibilities present themselves as protective measures against compromising emissions from data processing and data transmitting equipment. The use of shielding in the form of room shields, interlocking of supply lines, and filters for electric lines is the best protection for the user of data processing, data transmitting, and data handling equipment.