Back-of-Device (BoD) interfaces have emerged as a promising solution to free up screen real estate in smartphones by offloading

PIN authentication is the first line of defense for protecting private data on many smartphone applications, such as lock screens, messengers, and banking apps. However, existing PIN authentication systems have several constraints regarding security, usability, and robustness. To go beyond their limitations, this paper presents TaPIN, a reliable system that authenticates smartphone users with the collaborative use of PINs and tap biometrics. A user is first instructed to enter her PIN by tapping a smartphone screen for authentication. During the PIN entry, the user's fingertip collides with the screen, producing user-specific vibration and sound signals. TaPIN then senses the tap-induced signals and the collision properties, e.g., pressures and sizes, using the smartphone's built-in sensors and leverages them as biometric features. That is, it authenticates the user by verifying not only the entered PIN but also the collected features. Our experiments with 20 real-world users demonstrate that this two-factor authentication system is easy to use, more secure than existing methods, and deployable without dedicated hardware. For example, it accurately authenticates users with an average EER of 1.9% in stationary environments and maintains a reasonable level of security regardless of devices, tap styles, and noise.

Samsung Pay, a widely-used mobile payment service, enables users to pay using just their smartphone thanks to Magnetic Secure Transmission (MST). This technology facilitates communication between smartphones and magnetic card terminals by transmitting payment tokens through magnetic waves. Intriguingly, such magnetic waves inherently produce a distinct sound pattern (calledMST sound) containing payment information, which opens up new opportunities for both potential attackers and payment users. That is, MST sound can serve either as a new side channel for attackers to eavesdrop on MST transactions or as an easily accessible communication channel that enhances the payment experience for users. Inspired by these possibilities, we aim to deeply explore the potential of MST sound across these two dimensions, presenting two frameworks with different objectives: MagSnoop and M2APay. The first is the inference framework, which accurately, robustly, and efficiently infers payment tokens by listening to MST sounds. The second is the payment framework, which helps users establish a secure communication channel between MST-supported smartphones and microphone-equipped smartphones by shielding the vulnerability inherent in MST sound. Our experiments with prototypes of these frameworks achieved high accuracy in token inference and data transmission. Furthermore, both MagSnoop and M2APay are capable of accurately decoding tokens in diverse payment environments, including noisy environments and real-world scenarios.

Samsung Pay, one of the most representative mobile payment services, allows mobile users to make payment transactions almost anywhere using only their smartphone. This is thanks to MST (Magnetic Secure Transmission) that supports communication between smartphones and payment terminals for magnetic cards by transferring payment tokens via magnetic waves. Several attack methods have targeted this new technology by eavesdropping on magnetic fields to intercept the tokens, but with the use of dedicated hardware. This paper raises new security concerns for mobile payment users in a different, yet more effective way; by introducing MagSnoop, a novel framework that infers payment tokens from listening to MST sounds generated during the activation of MST payment transactions. More specifically, we first explore the principle, causing the generation of MST sounds, and the fundamental characteristics of these sounds. We then use these observations to infer payment tokens with a high degree of accuracy, robustness, applicability, and data efficiency. Our experiments with a prototype of MagSnoop demonstrate that it can support high accuracy in token inference (more than 77.8%). In addition, MagSnoop can maintain a reasonable level of accuracy regardless of the payment environments (e.g., 69.2% with a noise level of 50 dBA) and even in the real world (an inference success rate of 68.0% with 15 real-world users).

In this paper, we present TapAuth, a novel smartphone user authentication system that leverages vibrations generated from a user's tap inputs. TapAuth is based on the observation that the vibrations have their unique characteristic depending on the user due to differences in finger structure. Specifically, TapAuth collects audio and motion data using only built-in sensors, while users are entering PIN code. Then, it authenticates the user by verifying the validity of the code and comparing the collected data with a pre-built training dataset. Our experimental results with 20 real-world users show that TapAuth can achieve high accuracy.

Tap changing transformers are commonly used to reduce the imbalance in phase voltages or maintain the voltage magnitude in the system at a range, often 0.95 to 1.05 p.u. If an electric fault occurs on a power distribution network with such a tap-changing transformer, the tap adjusted on either the primary or secondary side changes the magnitude and angle of the short-circuit current. Therefore, power-flow analysis algorithms must be able to model tap-changing transformers. For example, several such algorithms (e.g., the Newton-Raphson, Gauss-Seidel, and fast decoupled methods that use the bus admittance matrix) require an inversion of a Jacobian matrix. If the matrix size is sufficiently large (e.g., up to thousands of dimensions), such methods may fail to calculate the inversion of a large matrix within a limited time despite the matrix reduction techniques. Moreover, if the admittance matrix of a tap-changing transformer is singular, its inverse matrix could not be found by the inversion of the matrix. The bus impedance matrix is commonly necessary for short-circuit studies. Thus, the singularity problem has been solved by a backward and forward sweep method. But, the method may not work for heavily-meshed distribution networks. Thus, this study presents a novel method that models tap-changing transformers, not causing the singularity. Then, the proposed method is verified in various case studies.

In this paper, we propose a novel segmentation method to extract business card region from the image. In our method, an input image is partitioned into four blocks and the probabilistic Hough transform is applied to each block to detect the line segment candidates of business card boundary. Then, our method searches the largest quadrilateral, which is formed by using the detected line segments, under some constraints through RANSAC-like method. To evaluate the proposed method, we test our method on the collected business card images having various kinds of backgrounds. As a result of experiments, we show that our method has achieved about segmentation rate of 90%.