SCREEN READING: ELECTROMAGNETIC INFORMATION LEAKAGE FROM THE COMPUTER MONITOR

variety of countermeasures, such as shielding, zoning, soft TEMPEST, and similar techniques, can be used to prevent data leakage.


Introduction
In recent years, new technologies have made it possible to exfiltrate sensitive data from a computer by monitoring the computer screen in a variety of novel ways that do not require network connectivity or physically contacting devices via the invisible channel determined by the computer screen. Because the user does not have a visual perception of what is happening, malware on the compromised computer can obtain sensitive data such as files, images, or passwords. The prevention of attacks using electromagnetic (EM) signals that are either conducted or radiated is referred to as emission security. By formulating that "changing electrical currents induce changing magnetic fields, which induce changing currents and induce a changing magnetic field that propagates as an EM wave through surrounding space," Oersted, Faraday, and Henry discovered the physics of EM emanation (Rowe, 2006). This field can be picked up by nearby electrical conductors and, through EM interference, can impede the operation of other electromagnetic devices. As a result, an antenna with an amplifier can pick up some signal from a computer and reconstruct generated electrical signals (Rowe, 2006). Military and commercial organizations are very concerned about the Transient Electro Magnetic Pulse Emanation Standard (TEMPEST) defence which prevents the stray EM pulses emitted by computers and other electronic devices from being picked up and used to reconstruct the sensitive data (Markagić, 2018, pp.143-153). TEMPEST has recently become a commercial issue for electronic voting machines and smart cards used for digital signatures. Side-channel attacks refer to a variety of attacks that take advantage of optical, thermal and acoustic emanations from the equipment. This happens when information leaks through a channel that is not intended for communication. Electromagnetic eavesdropping attacks can cause a computer to emit a stronger signal than usual and modulate the signal so that it can pass through the firewall.
Electromagnetic compatibility (EMC) and radio frequency interference (RFI) are closely related to EM security measures. All emission security issues are expected to worsen as more devices connect to wireless networks and processor speeds increase into the gigahertz range. There are two types of electromagnetic attacks that are not mutually exclusive: 1) when the signal is transmitted over a circuit such as a power line or phone line, it is known as Highjack and 2) when the signal is transmitted as radio frequency (RF) energy, it is known as TEMPEST. Properly shielded equipment is typically limited in quantity and designed specifically for defence markets, making it extremely expensive. The operating rooms must also be properly filtered.
Screen signals can be found in a variety of locations across computer networks. These signals may contain multiple harmonics, some of which radiate more effectively than others, owing to the designed equipment being certified to not emit any signals beyond a certain distance. Spying on the surface of a screen with a powerful telescope is a very basic approach to spying on the content displayed on it (Lavaud et al, 2021). Khun (2002), Backes et al (2008), and Backes et al (2009), on the other hand, describe several more efficient ways to attack computer monitor content. Computer monitors leak electromagnetic information as a result of three key factors used to reproduce video images: (1) refresh rate, (2) horizontal frequency, and (3) pixel frequency, which is the display principle (Mao et al, 2017). One method for estimating the risk of information leakage is to use multi-resolution spectrum analysis to distinguish and match the spectrum interval from the radiated EM signals.
This paper investigates the impact of how a side-channel attack causes compromised information to be taken from a computer screen. This paper also discusses the leakage of electromagnetic information from computer screens. To explain potential malicious attacks on computer monitors, software-defined radios (SDRs) are described.

Side-channel attacks
The security of a cryptosystem (cryptographic algorithms and protocols, cryptographic keys, and cryptographic devices used for implementation) is dependent on more than just using robust algorithms and parameters, certified protocols, and cryptographic keys that are long enough. Physical attacks on a system can also be used to compromise it. Side-channel attacks are generally physical attacks in which malicious parties extract confidential and protected data by observing how systems physically behave (Barthe et al, 2018). These attacks use the dependency between secret information used in the cryptosystem and physical values measured on/around the cryptosystem (e.g. power consumption, electromagnetic radiation, timing information) to break a system (Mangard et al, 2007). Table 1 depicts the classification of side-channel emanation (Lavaud et al, 2021). Each side-channel attack seeks to exploit an unintentional emission. As a result, the subject of side-channel attacks covers a broad range of techniques (Sayakkara et al, 2018). Side-channel information sources, such as EM emanations from a chip (Agrawal et al, 2003) and timings for various operations performed (Kocher, 1996, pp.104-113) have also been demonstrated to be exploitable (Mangard et al, 2007). Hayashi et al (2014, pp.954-965) conducted a thorough examination of EM emanations from a chip in-depth, including countermeasures. Their primary focus, however, was on recovering sensitive information from inside the computer systems (cryptographic keys, not-the-screen content). Kinugawa et al (2019, pp.62-90) demonstrate how to increase the EM leakage with a (cheap) hardware modification added to potentially any device and spread the attack over a greater distance. The authors show that the additional circuitry (interceptor) increases leakage and forces leakage in devices that are not susceptible to EM leakage.  Goller & Sigl (2015, pp.255-270) proposed to perform side-channel attacks on smartphones using standard radio equipment. The authors also show the ability to distinguish between squaring and multiplications. This discovery may result in the complete recovery of the Rivest, Shamir, and Adelman (RSA) key (Jonsson & Kaliski, 2003). Their setup gathered electromagnetic leaks from an Android phone. Genkin et al (2015, pp.95-112), and Genkin et al (2019, pp.853-869) present the extraction of cryptographic keys such as RSA or ElGamal from laptops using various side channels such as power and EM radiation (Will & Ko, 2015). Furthermore, an adversary may be able to monitor a device's power consumption while it performs secret key operations (Kocher et al, 2011, pp.5-27). Acoustic emanation from various computer system components can be used to exfiltrate data. Genkin et al (2014, pp.444-461) demonstrated that, by listening for acoustic emanation, it is possible to distinguish between CPU operations, resulting in an attack on an RSA algorithm encryption key. Fenkin et al (2019) show how to extract screen content using the acoustic side channel. Microphones can pick up sound from webcams or screens and transmit it during a video conference call or archived recordings. Berger et al (2006, pp.245-254) demonstrated a dictionary attack using keyboard acoustic emanation. Backes et al (2010) investigated acoustic side channels in printers. Asonov and Agrawal (2004) used the sound emitted by different keys to recover information typed on a keyboard. The contribution of Liu et al (2021, pp.1-15) is a sidechannel attack analysis that exploits the EM emanations of the display cable from a mobile phone. These signals are more difficult to obtain and may be significantly weaker than those examined in more traditional TEMPEST technique attacks. TEMPEST is a side-channel technique for spying on computer systems via unintentional radio or electrical signals, sounds, and vibrations (Kuhn & Anderson, 1998, pp.124-142). The possibility of intercepting visual information displayed on an electronic device screen is the most well-known issue associated with EM revealing emissions. Van Eck (1985, pp.269-286) is the first to present an unclassified analysis of the feasibility and security risks of computer monitor emanations. He was able to listen in on a real system from hundreds of meters away by measuring electromagnetic emanations with only $15 in equipment and a Cathode-Ray Tube (CRT) television set.
Side-channel attacks have a variety of countermeasures because they are among the most serious threats to embedded crypto devices and frequently target the secret (cryptographic) key in a device that secures sensitive data. The countermeasures' primary goal is to eliminate reliance on sensitive data and the side channel. One method attempts to separate the actual data processed by the device from the data on which the computation is performed (masking) (Prouff & Rivian, 2013, pp.142-159). Another approach attempts to separate the device's computed data from the power consumed by the computations (hiding). One of the countermeasures is also flattening the power consumption of a device. Hardware-based countermeasures propose microarchitecture-based solutions such as providing hardware support for advanced encryption standard (AES) instructions or making caches security-sensitive. Hardware countermeasures are effective, but they can be difficult to implement. In contrast, software countermeasures are simple to Electromagnetic information leakage from the computer monitor EM radiation is the underlying technology for wireless communication, and it is selected based on the distance to be covered, data throughput rate, signal frequency, amount of bandwidth required, modulation technique, power of the transmitted signal, and other factors (Sayakkara et al, 2018). Although wireless communication devices are designed to generate EM radiation at the appropriate frequency and amplitude for the communication technology, as a by-product of their internal operations, these devices also generate EM radiation at unintended frequencies (Genkin et al, 2014, pp.444-461). Unintentional EM emissions from computers can be caused by a variety of factors. The source of each EM signal determines the nature of these EM signals as well as the type of side-channel information they carry. The possibility of intercepting visual information displayed on computer monitors is the most well-known issue associated with the issue of EM revealing emissions. Van Eck (1985, pp.269-286) demonstrated a modified television set that was capable of capturing and visualizing video streams displayed on a nearby television screen. To transmit video data to computer monitors, various protocols are used, necessitating more flexibility than a dedicated hardware-based attack. This article was about CRT monitors. It should be noted that liquidcrystal displays (LCD), which are common output components of computers and currently dominate the market, are not immune to this threat because they are equipped with digital video data (DVD) transmission interfaces. This is not the case, because digital signals, like analogue signals, are susceptible to electromagnetic infiltration and enable non-invasive data acquisition. There is a risk of eavesdropping on the leaked signal because the leakage of the displayed information is quite high. In 2002, Kuhn expands on this eavesdropping concept by conducting an analysis of EM side-channel eavesdropping on modern video display technologies (Kuhn, 2002, pp.3-18). This study employs RF acquisition hardware with fast sampling rates to monitor EM emissions from computer displays. Sekiguchi (2010, pp.127-131)  monitor. Elibol et al (2012Elibol et al ( , pp.1767Elibol et al ( -1771) demonstrated a monitor eavesdropping system that remotely reconstructs screen images using RF acquisition hardware. The signal acquisition hardware is a portable platform that can operate at a variety of RF frequencies. In this work, the averaging of adjacent frames is used to improve the readability of the text. In 2016, Lee et al demonstrated the possibility of display information leakage by analysing electromagnetic emissions from desktop and laptop monitors (Lee et al, 2016). By analysing the display mechanism, the characteristics of the leaked signal from the LCD monitor are verified, and electromagnetic emanations are measured over a long distance using an eavesdropping experiment. Using a variety of signal processing techniques, the authors recovered display information.
Software-defined radio: How to spy on?
There have been practical challenges in the more demanding SDR applications, primarily due to analogue to digital conversion (ADC) and digital to analogue conversion (DAC) limitation trade-offs. Many of these compromises are being limited to higher frequencies due to faster ADC/DACs and higher resolution. To avoid being limited to singlefrequency ranges and to deal with multiple channels at once, SDR requires a wideband. Wideband performance is required to allow for dynamic spectrum and radio parameter management. The SDR should be able to digitize the desired frequency spectrum directly from an antenna, present it to a DSP processor, and output it to an application, as well as the reverse for a transmitter. The following benefits are provided by SDR: (1) flexibility, (2) interoperability, (3) ease of upgrade, (4) efficiency, and (5) higher-level interfaces. Figure 1 depicts the basic structure of an SDR receiver (Benks, 2016, pp.1-16).  (Benks, 2016) Рис. 1 -SDR приемник (Benks, 2016) Слика 1 -SDR пријемник The sampling rate of the RF acquisition hardware is the most important factor in the accuracy of the screen image reconstruction. SDRs provide greater flexibility at a lower cost, but their sampling rate is lower than that of dedicated RF acquisition hardware. Digital processors give radio equipment the flexibility of a programmable system, allowing a communication system to be changed simply by changing its software. Under the SDR paradigm, the task of configuring the radio's behaviour is transferred to software, leaving the hardware only to implement the radiofrequency front end. As a result, the radio is transformed into a dynamic element capable of changing its operational characteristics (bandwidth, modulation, coding rate) based on software configuration. The SDR is defined as "radio in which some or all physical layer functions are defined by software" (Garcia Reis et al, 2012). SDR devices use real-time software module execution on microprocessor platforms or digital signal processors, fast programmable gate arrays (FPGA) are commonly used for transmitting or receiving radio signals, the main operational characteristics of SDRs are modified at runtime, and the system can be easily reconfigured to perform different functions (Chamran et al, 2020). SDRs apply to a wide range of radiofrequency technologies, and their standards have made base station software updates more appealing than costly base station replacements. The SDR expands the possibilities by making it easier to implement existing radio applications and enabling new types of applications. The availability of low-cost devices that receive and digitize radiofrequency signals has brought the SDR to both professional and home engineering desks (Stewart et al, 2015, pp.64-71). In their work, Molina-Tenorio et al (2021, pp.1-21) describe the characteristics of the SDR-RTL (Nooelec, 2021), HackRF One (Great scott gadgets, 2021), and LimeSDR Mini (Lime microsystems, 2021) devices. Table 2 lists the main characteristics of these devices. Rugeles Uribe et al (2021, pp.1-13) compare 19 more commercially available SDR platforms in terms of ADC/DAC, Tx/Rx, Fmin-Fmax and Max RF Bandwidth, all of which were collected in 2019 and 2020: FUNcube Dongle, RSPduo, Airspy-mini, Airspy-R2, Pluto, BladeRF 2.0 Micro, AD-FMCOMMS4-EBZ, USRP-1, PicoSDR, WARP-V3, USRP-N210, TMDSSFFSDR, USRP X310, USRP-2974, AIR-T, Sidekiq X4, USRP N320 and CRIMSON Cyan, all of them being commercially available SDR platforms. The authors also discuss how hardware is evolving to increase computational capacity. Furthermore, the authors present the Bastille Network classification and descriptions of wireless vulnerabilities (Bastille Networks, 2020).
Electromagnetic compatibility standards and regulations in consumer products address emerging threats from eavesdropping attacks using EM side-channels. According to the International Organization for Standardisation (ISO) ITU-T advisory notice K.841, when considering the EMC requirements of consumer devices, information leakage from EM emissions must be considered (ITU, 2014). Many hardware-software tools, however, use SDR platforms for side-channel attacks on computer monitors. TempestSDR, an open-source software library that uses SDR platforms for EM side-channel attacks on computer monitors, is one of the most well-known. It is capable of automatically detecting the dimensions and frame rate of a target when the target monitor details are unknown by identifying repeating patterns in the EM signal that correspond to the individual frames of the video (see Figure 2). TempestSDR allows the user to use any SDR that supports ExtIO (such as those described in Table 2) to receive unintentional signal radiation from the screen and convert that signal back into a live image, allowing them to see what is on screen without a direct connection (RTL-SDR, 2017).
How to defend against computer monitor attacks?
Electronic security protection, like all security measures, must be cost-effective, that is, it must eliminate security issues without interfering with system performance (Doychev, 2016;Rowe, 2006). The sources of unwanted signals must be protected by implementing solutions that effectively prevent the infiltration process from taking place (Levina et al, 2019, pp.393-400;Ometov et al, 2017Ometov et al, , pp.2591Ometov et al, -2601. Rowe summarizes the suitability of various electronic security methods for mitigating various threats. Table 3 shows the results for the monitor, power, and cables. It is essential to emphasize the significance of differences in military and civil security standards. In 2006, Khun predicted that simple eavesdropping tools for compromising emanations would soon be available for free download from the Internet. For today's information security professionals, the question is: What protective countermeasures are available for computers that display extremely sensitive data regularly? These could be jamming devices used to intentionally increase environmental noise, metal shielding, zoning, soft TEMPEST, and other similar techniques. Today, jamming devices are rarely used because the jamming signal must be carefully selected and synchronized with the signal to be covered, and jamming devices may draw the attention of eavesdroppers to the location of equipment. The EM shielding protects devices, cables and rooms against compromising emanations ( Kuhn, 2006, pp.1-10). To eliminate emanations, the source of the emanations should be placed in metal boxes made of conductive materials (copper, aluminium, steel), also known as Faraday cages. However, perfect protection requires that the conductive enclosure remain intact. Because gaps are required for ventilation, power lines, keyboards, and network connections, these gaps may allow signals to leak out (Warne & Chen, 1992, pp.173-182). Molyneux-Child (1997) recommends at least one-tenth to prevent significant radiation from escaping and one-hundredth to provide a 60 dB reduction. Creating meandering channels through the gaps, as well as using waveguides in the form of conductive pipes through the gaps, can help to reduce the emanations at these gaps. Power lines can be filtered through these gaps, and fibfibretic cables can transmit data without requiring an electromagnetic channel. A conductive film can be applied to monitor screens, but keyboards are more difficult to protect. Due to the difficulty of shielding, both devices and their locations can be classified to indicate how close an eavesdropper can get (Zone 0: eavesdropper can be within 1-20 m; Zone 1: eavesdropper can be within 20-100 m; Zone 2: eavesdropper can be 100 m to 1 km; Zone 3: eavesdropper cannot be closer than a kilometre). These measurements are based on the assumption that only space exists between the eavesdropper and the target. Przybysz et al (2021, pp.1-15) discuss publicly available fragments of the American military requirements NSTISSAM TEMPEST/1-92 (Cryptome, 2008) and NSTISSAM TEMPEST/2-95 (Cryptome, 2000), which define three levels of security for devices that could be used in information processing zones and whose R rays meet the following conditions: R 20 m, R 100 m, and R > 100 m. (Emission levels must be measured from a distance of one meter). The MIL-STD-461G document (EverySpec, 2015) also recommends this measurement distance. The authors concluded promising emission signals emitted by commercial devices can be detected from a few dozen meters away. De Meulemeester et al (2020) confirmed this by demonstrating that visual information could be recovered from a distance of approximately 80 m. Various software countermeasures, such as deliberate softening of font edges or randomization of less significant bits in the frame buffer, can be used to protect against EM leakage from the computer monitor (Duc et al, 2019(Duc et al, , pp.1263(Duc et al, -1297. Safe fonts are one of the security measures developed using Kubiak's safety criteria (Kubiak, 2020), and they have the following characteristics: (1) The lines that form the characters intersect at right angles, implying that each character is made entirely of vertical and horizontal lines, (2) font characters are devoid of decorative and diagonal

Conclusion
New technologies enable malicious scanning of information emitted by the computer monitor. The TEMPEST defence, which prevents stray EM pulses emitted by computers and other electronic devices from being picked up and used to reconstruct sensitive data, is causing concern among military and commercial organizations. There are two types of EM attacks: highjack and RF energy attacks. Screen signals contain multiple harmonics, some of which radiate more effectively than others. The refresh rate, horizontal frequency, and pixel frequency of computer monitors all leak electromagnetic information. This paper describes how a sidechannel attack causes compromised information to be taken from a computer screen. The SDR is used to explain how visual data can be intercepted. To describe the possibility of protecting the data radiated from the computer monitor, a variety of countermeasures such as shielding, zoning, soft TEMPEST, and similar techniques are described.