• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.4
• /
• pp.803-812
• /
• 2018

The Morse code was firstly used for a wire telegraph service on May 1884 in USA. After about 10 years Korean researchers developed the Morse code in 1884 and the Morse code was firstly introduced in Korea for the wire telegraph service in September 1885. For about 130 years later, the Hangul Jamo Morse code has been used without any serious modification up to now. The Morse code is the variable length code system which each codes have a different code length. The related reference shows that English alphabet Morse code was applied in the coding theory. However, a result of analysis showed that the Hangul Jamo Morse code was not applied in the coding theory as the english alphabet code system. This paper presents the analysis result of Hangul Jamo Morse code according to the variable length coding theory. As a result of the analysis, when Hangul Jamo are coded with application of the variable length coding theory, the transmission data rate could be increased up to 25% compare to the existing code system. This paper shows a new Hangul Jamo Morse code system based on the analysis result.

• Journal of KIISE
• /
• v.43 no.11
• /
• pp.1275-1280
• /
• 2016

Existing techniques for string similarity search first generate a set of candidate strings and then verify the candidates. The efficiency of string similarity search is highly dependent on candidate generation methods. State of the art techniques select fixed length q-grams from a query string and generate candidates using inverted lists of the selected q-grams. In this paper, we propose a technique to generate candidates using variable length grams of a query string and develop a dynamic programming algorithm that selects an optimal combination of variable length grams from a query string. Experimental results show that the proposed technique improves the performance of string similarity search compared with the existing techniques.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.319-325
• /
• 2022

The variation of utterance lengths is a representative factor that can degrade the performance of speaker verification systems. To handle this issue, previous studies had attempted to extract speaker features from various branches or to use convolution layers with different receptive fields. Combining the advantages of the previous two approaches for variable-length input, this paper proposes integrated receptive field diversification that extracts speaker features through more diverse receptive field. The proposed method processes the input features by convolutional layers with different receptive fields at multiple time-axis branches, and extracts speaker embedding by dynamically aggregating the processed features according to the lengths of input utterances. The deep neural networks in this study were trained on the VoxCeleb2 dataset and tested on the VoxCeleb1 evaluation dataset that divided into 1 s, 2 s, 5 s, and full-length. Experimental results demonstrated that the proposed method reduces the equal error rate by 19.7 % compared to the baseline.

• Proceedings of the Korea Multimedia Society Conference
• /
• 1998.04a
• /
• pp.216-221
• /
• 1998

본 논문에서는 새로운 가변길이 블록 암호알고리즘을 제안한다. 제안된 블록 암호알고리즘은 128비트에서 256비트까지의 가변적인 키 길이를 가지며 가변적인 라운드 수를 사용한다. 각 라운드는 서로 다른 두 개의 F함수를 사용하여 2단계로 구성되는 double involution 구조를 사용하며, 또한 두 개의 서로 다른 키 스케줄링 알고리즘을 사용하여 알려진 공격에 대해 안전하도록 설계하였다.

• Journal of Korea Multimedia Society
• /
• v.1 no.1
• /
• pp.90-97
• /
• 1998

In this paper, we propose a new variable length block cipher. It has a variable key length from 128-bit to 256-bit and uses a variable number of rounds. In each round, the proposed algorithm uses the double involution structure which consists of tow steps and two different F functions. In addition, the proposed algorithm has two different key schedulings for providing the strength against known attacks.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.750-758
• /
• 2016

In general, a system can be stable when it is designed with a rigid material. However, the use of a rigid system can be limited, such as grasping a glass or using a small surgical instrument. To resolve this limitation, a variable stiffness mechanism was developed using a flexible material. Previous research verified the variable stiffness mechanism where flexible segments and rigid segments were connected alternately in series. However, research into the design parameters of the variable stiffness structure is needed to satisfy the desired stiffness. Therefore, a variable stiffness structure was tested by varying the design parameters to confirm the trend of the stiffness variation. When the radius of the structure becomes larger, the stiffness increases. The stiffness increased with decreasing length of the flexible segments. Under the same design parameters, the length of the flexible segments had a greater effect on the stiffness than the length of the rigid segments. In addition, the stiffness was estimated using the pseudo rigid body model and was compared with the experimental results. This parametric study can be used as a design guideline for designing the variable stiffness mechanism to satisfy the desired stiffness.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.16-17
• /
• 2016

일반적인 에디터 프로그램과 운영체제를 사용하여 데이터를 편집할 경우, 일부 수정에도 모든 데이터를 다시 저장하고 있다. 본 논문에서는 이러한 기존 파일시스템의 문제점을 개선하기 위해 가변 길이 블록의 개념을 커널 레벨에 적용하여 효율적으로 수정된 데이터를 다시 쓸 수 있도록 하였다. 가변 블록은 데이터 삭제가 발생 했을 경우 나머지 블록은 그대로 유지하고 수정된 블록만 다시 저장하며, 삭제된 데이터로 인해 생기는 빈 공간의 그대로 유지하는 방법을 말한다. 이 개념을 기존 리눅스에서 사용하는 ext4파일 시스템에 적용하여 시스템을 구축, 실험하였으며 결과적으로 쓰기 연산을 비롯해 CPU사용량에서 크게 성능을 향상시켰다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.04a
• /
• pp.51-53
• /
• 2015

본 논문은 기존의 파일 시스템에서 파일의 일부를 수정했을 경우 수정된 위치 이후의 모든 블록이 수정되는 문제점을 개선하고자 가변 길이 블록 파일 시스템을 제안한다. 가변 길이 블록 파일 시스템은 파일 수정 시 수정이 발생한 데이터 블록만 새로 저장하고 나머지 블록의 상태는 유지함으로써 파일의 쓰기 연산을 최소화시킨다. 제안된 시스템을 사용할 경우 파일 수정 시 기존 파일 시스템보다 빠른 속도로 파일의 내용을 변경 시킬 수 있으며, 특히 대용량 파일에서 우수한 성능을 보인다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.3
• /
• pp.187-187
• /
• 2004

In this paper, we present an efficient architecture of a versatile Reed-Solomon (RS) decoder which can be programmed to decode RS codes continuously with my message length k as well as any block length n. This unique feature eliminates the need of inserting zeros for decoding shortened RS codes. Also, the values of the parameters n and k, hence the error-correcting capability t can be altered at every codeword block. The decoder permits 3-step pipelined processing based on the modified Euclid's algorithm (MEA). Since each step can be driven by a separate clock, the decoder can operate just as 2-step pipeline processing by employing the faster clock in step 2 and/or step 3. Also, the decoder can be used even in the case that the input clock is different from the output clock. Each step is designed to have a structure suitable for decoding RS codes with varying block length. A new architecture for the MEA is designed for variable values of the t. The operating length of the shift registers in the MEA block is shortened by one, and it can be varied according to the different values of the t. To maintain the throughput rate with less circuitry, the MEA block uses both the recursive technique and the over-clocking technique. The decoder can decodes codeword received not only in a burst mode, but also in a continuous mode. It can be used in a wide range of applications because of its versatility. The adaptive RS decoder over GF($2^8$) having the error-correcting capability of upto 10 has been designed in VHDL, and successfully synthesized in an FPGA chip.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.3
• /
• pp.29-38
• /
• 2004

In this paper, we present an efficient architecture of a versatile Reed-Solomon (RS) decoder which can be programmed to decode RS codes continuously with my message length k as well as any block length n. This unique feature eliminates the need of inserting zeros for decoding shortened RS codes. Also, the values of the parameters n and k, hence the error-correcting capability t can be altered at every codeword block. The decoder permits 3-step pipelined processing based on the modified Euclid's algorithm (MEA). Since each step can be driven by a separate clock, the decoder can operate just as 2-step pipeline processing by employing the faster clock in step 2 and/or step 3. Also, the decoder can be used even in the case that the input clock is different from the output clock. Each step is designed to have a structure suitable for decoding RS codes with varying block length. A new architecture for the MEA is designed for variable values of the t. The operating length of the shift registers in the MEA block is shortened by one, and it can be varied according to the different values of the t. To maintain the throughput rate with less circuitry, the MEA block uses both the recursive technique and the over-clocking technique. The decoder can decodes codeword received not only in a burst mode, but also in a continuous mode. It can be used in a wide range of applications because of its versatility. The adaptive RS decoder over GF(2$^{8}$ ) having the error-correcting capability of upto 10 has been designed in VHDL, and successfully synthesized in an FPGA chip.