Video Sequence Segmentation Based on K-Means in Air-Gap Data Transmission for a Cluttered Environment

Abstract

An air gap is a technique that increases the security of information systems. The use of unconventional communication channels allows for obtaining communication that is of interest to the attacker as well as to cybersecurity engineers. One of the very dangerous forms of attack is the use of computer screen brightness modulation, which is not visible to the user but can be observed from a distance by the attacker. Once infected, the computer can transmit data over long distances. Even in the absence of direct screen visibility, transmission can be realized by analyzing the modulated reflection of the monitor’s afterglow. The paper presents a new method for the automatic segmentation of video sequences to retrieve the transmitted data that does not have the drawbacks of the heretofore known method of growth (filling) based on an analysis of adjacent pixels. A fast camera operating at 380 fps was used for image acquisition. The method uses the characteristics of the amplitude spectrum for individual pixels, which is specific to the light sources in the room, and clustering with the k-means algorithm to group pixels into larger areas. Then, using the averaging of values for individual areas, it is possible to recover the 2-PAM (pulse-amplitude modulation) signal even at a 1000 times greater level of interference in the area to the transmitted signal, as shown in the experiments. The method does not require high-quality lenses.

Insight into Anomaly Detection and Prediction and Mobile Network Security Enhancement Leveraging K-Means Clustering on Call Detail Records

“…K-means clustering is a widely used unsupervised machine learning algorithm that partitions data points into distinct clusters based on their similarities [ 5 ]. However, there has been negligible prior investigation into anomaly detection, specifically in the context of utilizing CDR data for predicting and forecasting anomalies to enhance mobile network security.…”