• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.34 no.6
• /
• pp.63-69
• /
• 2018

The machine learning model can capture the dynamics of building systems with less inputs than the first principle based simulation model. The training data for developing a machine learning model are usually selected in a heuristic manner. In this study, the authors developed a machine learning model which can describe supply air temperature from an AHU in a real office building. For rational reduction of the training data, the progressive sampling method was used. It is found that even though the progressive sampling requires far less training data (n=60) than the offline regular sampling (n=1,799), the MBEs of both models are similar (2.6% vs. 5.4%). In addition, for the update of the machine learning model, the normalized mutual information (NMI) was applied. If the NMI between the simulation output and the measured data is less than 0.2, the model has to be updated. By the use of the NMI, the model can perform better prediction ($5.4%{\rightarrow}1.3%$).

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.12 no.4
• /
• pp.77-85
• /
• 2002

We present the impossible differential cryptanalysis of the block cipher XTEA[7] and TEA[6]. The core of the design principle of these block ciphers is an easy implementation and a simplicity. But this simplicity dose not offer a large diffusion property. Our impossible differential cryptanalysis of reduced-round versions of XTEA and TEA is based on this fact. We will show how to construct a 12-round impossible characteristic of XTEA. We can then derive 128-bit user key of the 14-round XTEA with $2^{62.5}$ chosen plaintexts and $2^{85}$ encryption times using the 12-round impossible characteristic. In addition, we will show how to construct a 10-round impossible characteristic or TEA. Then we can derive 128-bit user key or the 11-round TEA with $2^{52.5}$ chosen plaintexts and $2^{84}$ encryption times using the 10-round impossible characteristic.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.1
• /
• pp.19-26
• /
• 2003

Recently, the number of internet service companies is increasing and so is the number of malicious attackers. Damage such as distrust about credit and instability of the service by these attacks may influence us fatally as it makes companies image failing down. One of the frequent and fatal attacks is DoS(Denial-of-Service). Because the attacker performs IP spoofing for hiding his location in DoS attack it is hard to get an exact location of the attacker from source IP address only. and even if the system recovers from the attack successfully, if attack origin has not been identified, we have to consider the possibility that there may be another attack again in near future by the same attacker. This study suggests to find the attack origin through MAC address marking of the attack origin. It is based on an IP trace algorithm, called Marking Algorithm. It modifies the Martins Algorithm so that we can convey the MAC address of the intervening routers, and as a result it can trace the exact IP address of the original attacker. To improve the detection time, our algorithm also contains a technique to improve the packet arrival rate. By adjusting marking probability according to the distance from the packet origin we were able to decrease the number of needed packets to traceback the IP address.

• Korea and Global Affairs
• /
• v.1 no.1
• /
• pp.31-70
• /
• 2017

On December 9, 2016, the decision of impeachment of the National Assembly decided against Park Geun-hye came from the nomination conflicts of the 20th general election between Pro-Park group and Anti-Park group at the ruling Saenuri Party. Therefore, this study focused on the nomination conflicts of the ruling party and the opposition party on the election results in the 20th general election on the public sentiment of the people. The nomination conflicts of the ruling and opposite parties were a prelude to the victory of the 19th presidential election, and it was serious faction conflict. Firstly, the study examined how the nomination conflicts of each party were centered on President Park Geun-hye as well as the leaders of each party and the chairman of the nomination committee. Secondly, the study examined what kind of changes would be made to the composition of presidential candidates for each party at the time of the presidential election. Thirdly, the study examined the opposition parties' separation between the Minjoo Party of Korea and the People's Party of Korea before and after the election and the issue of initiative in Honam. As a result of the analysis, the 20th general election failed to obtain a majority seat of the ruling Saenuri Party, and the opposition won and formed the majority. The reason why President Park and Saenuri were greatly defeated in the contest even in the situation where the opposition parties were divided is the root cause in the attitude of Pro-Park group and Anti-Park group who assumed their victory. Therefore, it is highly possible to render its responsibility to President Park Geun-hye, who has become a 'past power', and it has opened up the possibility that the emergence of future power by opposition parties. In the case of the opposition party, it is clear that the battle for Honam, which is a traditional opposition party's support group, is a matter of good fortune of the two major powers, Moon Jae In and Ahn Cheol Soo.