A Support Vector Machine for Identification of Monitors Based on Their Unintended Electromagnetic Emanation

Abstract

Abstract-Electrical equipments usually radiate unintended emission which carries characteristic information when running, such as emanation from computers monitors, keyboards and other components, this emanation can be possibly used to reconstruct the source information. Most of the experiments related to this area are carried out inside a semi-anechoic chamber, and measurement out of it may not be considered to be optimal, because the data captured are usually not sufficient. Yet in this study, we take LCD monitors as typical examples and find that characteristics significantly differ between products, parameters such as the magnitude and spectrum were measured under normal environment. We take the PCB traces as antennas and acquire the raw signal directly near the antenna and extract the parameters to use as input to support vector machine (SVM) which was trained to identify the emanating source (LCD monitors). In this study, the method was tested using the emission captured from one Samsung (SyncMaster E1920) and two LG (L1753s) monitors, and a laptop (ACER Aspire 5542). The SVM was able to classify the source of signals with 98.9510% accuracy while using emission that captured from the running monitors.

Eavesdropping on Black-box Mobile Devices via Audio Amplifier's EMR

“…It attracts many researchers' attentions of investigating the system vulnerabilities caused by leaking EMRs. Following this line of researches, EMRs have been exploited to infer victim's keystrokes on keyboards [24], [25], [43], profile device memory usages [44], identify the model of LCD monitors [34], recover the displayed information on mobile screens [35], and exfiltrate secret data by establishing electromagnetic covert channel [45]- [47]. In addition, the most recent studies show that fine-grained data processed on the device also have the leakage threats via EMRs.…”

“…As established in our feasibility study for answering RQ1&2, the EMRs radiated by mobile devices' audio amplifiers are primarily correlated with the original audio sounds and share a unique EMR spectrum characteristic for all amplifiers. Therefore, unlike previous EMR-based side-channel attacks [21], [22], [34], [35], we do not need knowledge of the device's hardware specifics or EMR characteristics to carry out our attack. Our focus, therefore, is to develop a training-free audio recovery scheme based solely on signal processing techniques.…”

Electromagnetic wave pattern detection using cepstral features in the manufacturing field

“…In the wireless communication fields, cepstral features are used to analyze the channel utilization state of primary users in [7] and identify modulation technique [8] but there are few examples using cepstral features for detection of electromagnetic noise. Focusing on features for detection of noise and signal, if data is investigated previously and information about characteristics of frequency is not needed, only using average and deviation values of power spectrum will work [9]. But features including information about spectrum pattern should be used when analyzing data containing signals which have obvious characteristics related to frequency such as narrow band communications.…”

Performance parameters prediction of slotted microstrip antennas with modified ground plane using support vector machine

“…Recently, the SVM concept has been introduced to predict the specific model accurately and instantly in several cases [16–31]. These cases are mentioned as: characterization of communication networks [16], annual runoff forecasting [17], blind multiuser detector for chaos-based CDMA system [18], building of synthesis of transmission line for microwave-integrated circuit [19], dielectric target detection through wall [20], urban impervious surface estimation from RADARSAT-2 Polari metric data [21], integrating a grid scheme (GS) into a least-squares support vector machine (LSSVM) with a mixed kernel to solve a data classification problem [22], estimating highly selective channels for Orthogonal Frequency Division Multiplexing system by complex LSSVM [23], calibration for position sensor [24], identify monitors on the basis of their unintended electromagnetic radiation [25], electromechanical coupling for microwave filter tuning [26], and detection and delineation of P- and T-wave in Electrocardiogram signals [27]. Consequently, different types of problems have been resolved using the formulation of SVMs but unfortunately, the literature of SVMs formulation in electromagnetic and microstrip antennas problems is very much limited [28–31].…”