• Journal of Multimedia Information System
• /
• v.9 no.2
• /
• pp.103-112
• /
• 2022

Background: A presentation attack places the printed image or displayed video at the front of the sensor to deceive the biometric recognition system. Usually, presentation attackers steal a genuine user's biometric image and use it for presentation attack. In recent years, reconstruction attack and adversarial attack can generate high-quality fake images, and have high attack success rates. However, their attack rates degrade remarkably after image shooting. Methods: In order to comprehensively analyze the threat of presentation attack to palmprint recognition system, this paper makes six palmprint presentation attack datasets. The datasets were tested on texture coding-based recognition methods and deep learning-based recognition methods. Results and conclusion: The experimental results show that the presentation attack caused by the leakage of the original image has a high success rate and a great threat; while the success rates of reconstruction attack and adversarial attack decrease significantly.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.55-62
• /
• 2018

The MTD-based approach changes various operating parameters dynamically so that the vulnerability of the system can be protected from the malicious attack. In this paper, random/serial scanning/jamming attack success probabilities have been mathematically analyzed and verified through simulation to improve the security of the wireless communication systems in which the MTD-SDR technologies are applied. As a result, for random scanning attacks, attack success probability increases as the change period of transmission channel increases, while for random jamming attacks there is no change. The attack success probability patterns for serial attacks are similar to those of random attacks, but when the change period of transmission channel approaches to the total number of transmission channels, the success probability of serial attack is getting greater than that of random attack, up to twice in jamming attacks and up to 36% in scanning attacks.

• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.878-885
• /
• 2022

Research in the field of computer vision based on deep learning is being actively conducted. However, deep learning-based models have vulnerabilities in adversarial attacks that increase the model's misclassification rate by applying adversarial perturbation. In particular, in the case of FGSM, it is recognized as one of the effective attack methods because it is simple, fast and has a considerable attack success rate. Meanwhile, as one of the efforts to visualize deep learning models, Grad-CAM enables visual explanation of convolutional neural networks. In this paper, I propose a method to generate adversarial examples with high attack success rate by applying Grad-CAM to FGSM. The method chooses fixels, which are closely related to labels, by using Grad-CAM and add perturbations to the fixels intensively. The proposed method has a higher success rate than the FGSM model in the same perturbation for both targeted and untargeted examples. In addition, unlike FGSM, it has the advantage that the distribution of noise is not uniform, and when the success rate is increased by repeatedly applying noise, the attack is successful with fewer iterations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.9
• /
• pp.4548-4559
• /
• 2018

Conditional differential attack against NFSR-based cryptosystems proposed by Knellwolf et al. in Asiacrypt 2010 has been widely used for analyzing round-reduced Grain v1. In this paper, we present improved conditional differential attacks on Grain v1 based on a factorization simplification method, which makes it possible to obtain the expressions of internal states in more rounds and analyze the expressions more precisely. Following a condition-imposing strategy that saves more IV bits, Sarkar's distinguishing attack on Grain v1 of 106 rounds is improved to a key recovery attack. Moreover, we show new distinguishing attack and key recovery attack on Grain v1 of 107 rounds with lower complexity O($2^{34}$) and appreciable theoretical success probability 93.7%. Most importantly, our attacks can practically recover key expressions with higher success probability than theoretical results.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.3
• /
• pp.125-132
• /
• 2013

In this paper, We send DDoS(Distributed Denial of Service) attack traffic to real homepages in real networks. We analyze the results of DDoS attack and propose mitigation method against DDoS Attacks. In order to analyze the results of DDoS Attacks, We group three defense level by administrative subjects: Top level defense, Middle level defense, Bottom level defense. Also We group four attack methods by feature. We describe the results that average of attack success rate on defense level and average of attack success rate on attack categories about 48ea homepages and 2ea exceptional cases. Finally, We propose mitigation method against DDoS attack.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.965-973
• /
• 2022

As IoT(Internet of Things) devices become common, many algorithms have been developed to protect users' personal information. The laser fault injection attack that threatens those algorithms is a side-channel analysis that intentionally injects a laser beam to the outside of a device to acquire confidential information or abnormal privileges of the system. There are many studies to determine the timing of fault injection to reduce the number of necessary fault injections, but the location to inject faults is only repeatedly searched for the entire area of the device. However, when fault injection is performed in an algorithm-independent area, the attacker cannot obtain the intended faulted statement or attempt to bypass authentication, so finding areas vulnerable to fault injection and performing an attack is an important consideration in achieving a high attack success rate. In this paper, we show that a 100% attack success rate can be achieved by determining the vulnerable areas for fault injection by using electromagnetic and thermal information generated from the device's chip. Based on this, we propose an efficient fault injection attack system.

• Korean Journal of Applied Biomechanics
• /
• v.15 no.1
• /
• pp.1-18
• /
• 2005

The purpose of this study was to evaluate performance characteristics. in the success of team attack during volleyball competition. For this, real game was analyzed using S-VHS video camera for three dimensional cinematography. The contents of this research were receive to set time, set to spike time, setter moving distance, ball position during set and attacker's relative ball position during jump and spike, and open area at spike. Among the various kinematical factors considered in this study, the ball height relative spiker and open area were found to be related to offensive performance. This finding implies that although their direct impact were not significant the other factors could have indirect impact on offensive performance by increasing open area. Thus, receiver, setter, assistant attacker, and main attacker should practice for successful performance in each stage. Lastly, it is necessary to develop kinematic variables to evaluate performance characteristics of players. Further study may consider the best defense position against the attack of the opponent player.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.3
• /
• pp.49-57
• /
• 2010

We extended a general attack tree to apply cyber attack model for network vulnerability analysis. We defined an extended cyber attack tree (E-CAT) which extends the general attack tree by associating each node of the tree with a transition of attack that could have contributed to the cyber attack. The E-CAT resolved the limitation that a general attack tree can not express complex and sophisticate attacks. Firstly, the Boolean expression can simply express attack scenario with symbols and codes. Secondary, An Attack Generation Probability is used to select attack method in an attack tree. A CONDITION-composition can express new and modified attack transition which a aeneral attack tree can not express. The E-CAT is possible to have attack's flexibility and improve attack success rate when it is applied to cyber attack model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1802-1819
• /
• 2017

Correlation-enhanced collision attack has been proposed by Moradi et al. for several years. However, in practical operations, this method costs lots of time on trace acquisition, storage and averaging due to its bytewise collision detection. In this paper, we propose a bitwise collision attack based on second-order distance model. In this method, only 9 average traces are enough to finish a collision attack. Furthermore, two candidate models are given in this study to distinguish collisions, and the corresponding practical experiments are also performed. The experimental results indicate that the operation time of our attack is only 8% of that of correlation-enhanced collision attack, when the two success rates are both above 0.9.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.1
• /
• pp.296-308
• /
• 2015

Advanced Encryption Standard (AES) is widely used for protecting wireless sensor network (WSN). At the Workshop on Cryptographic Hardware and Embedded Systems (CHES) 2012, G$\acute{e}$rard et al. proposed an optimized collision attack and break a practical implementation of AES. However, the attack needs at least 256 averaged power traces and has a high computational complexity because of its byte wise operation. In this paper, we propose a novel double sieve collision attack based on bitwise collision detection, and an improved version with an error-tolerant mechanism. Practical attacks are successfully conducted on a software implementation of AES in a low-power chip which can be used in wireless sensor node. Simulation results show that our attack needs 90% less time than the work published by G$\acute{e}$rard et al. to reach a success rate of 0.9.