• The Journal of the Acoustical Society of Korea
• /
• v.21 no.2
• /
• pp.190-198
• /
• 2002

In general, we have used the Viterbi algorithm of Speech recognition for decoding. But a decoder in speaker verification has to recognize same word of every speaker differently. In this paper, we propose a noble decoding algorithm that could replace the typical Viterbi algorithm for the speaker verification system. We utilize for the proposed algorithm the speaker adaptation algorithms that transform feature vectors into the region of the client' characteristics in the speech recognition. There are many adaptation algorithms, but we take MLLR (Maximum Likelihood Linear Regression) and MAP (Maximum A-Posterior) adaptation algorithms for proposed algorithm. We could achieve improvement of performance about 30% of EER (Equal Error Rate) using proposed algorithm instead of the typical Viterbi algorithm.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.4
• /
• pp.721-726
• /
• 2012

In this paper, we proposed an efficient algorithm based on SPFA(shortest path faster algorithm), which is an improved the Bellman-Ford algorithm. The Bellman-Ford algorithm can be used on graphs with negative edge weights unlike Dijkstra's algorithm. And SPFA algorithm used a queue to store the nodes, to avoid redundancy, though the Bellman-Ford algorithm takes a long time to update the nodes table. In this improved algorithm, an adjacency list is also used to store each vertex of the graph, applying dynamic optimal approach. And a queue is used to store the data. The improved algorithm can find the optimal path by continuous relaxation operation to the new node. Simulations to compare the efficiencies for Dijkstra's algorithm, SPFA algorithm and improved Bellman-Ford were taken. The result shows that Dijkstra's algorithm, SPFA algorithm have almost same efficiency on the random graphs, the improved algorithm, although the improved algorithm is not desirable, on grid maps the proposed algorithm is very efficient. The proposed algorithm has reduced two-third times processing time than SPFA algorithm.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.13 no.2
• /
• pp.130-135
• /
• 2002

A new Sign algorithm which has improved convergence speed is presented. The data-recycling technique, whose coefficients are multiply adapted in a symbol time period by recycling the received data, is applied to Sign algorithm which has few multiplications. Sign algorithm has very few multiplications and is the most easily implemented, but it gives small rate of convergence relative to others. The proposed algorithm combines the advatage of Sign algorithm, few multiplications, and the virtue of Data-Recycling LMS algorithm, simplicity and fast convergence. The results of computer simulation show that the proposed algorithm has 2 times faster convergence rate than that of LMS algorithm. Comparing to Data-Recycling LMS algorithm, in similar convergence conditions, it requires half fewer multiplications.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 1998.12a
• /
• pp.289-301
• /
• 1998

이동통신에서의 사용자 인증 서비스는 통화도용 방지와 신뢰성 있는 과금을 위한 중요한 보호 서비스이다. 본 논문에서는 IS-95A 인증 시스템에 적용 가능한 안전한 인증 알고리듬과 인증 키 생성 알고리듬을 제안한다. 특히 인증 알고리듬을 활용하여 인증 키 생성알고리듬을 Oracle 해쉬함수의 형태로 구현함으로써 인증 시스템의 높은 안전성과 간결성을 동시에 성취하였다. 그리고 통계적 분석 기법을 사용하여 개발된 알고리듬의 출력 특성을 분석한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.7
• /
• pp.1259-1265
• /
• 2006

In this paper, we propose a combined synthetic algorithm of the logic level for high speed FPGA design. The algorithm divides critical path to reduce delay time and generates a circuit which the divided circuits execute simultaneously. This kernel selection algorithm is made by C-langage of SUN UNIX. We compare this with the existing FlowMap algorithm. This proposed algorithm shows result on 33.3% reduction of delay time by comparison with the existing algorithm.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.181-182
• /
• 2008

본 논문에서는 연산 최적화 알고리듬 중 PSO(Particle Swarm Optimization) 알고리듬을 NVIDIA사(社)에서 제공한 CUDA(Compute Unified Device Architecture)를 이용하여 새롭게 구현하였다. CUDA는 CPU가 아닌 GPU(Graphic Processing Unit)의 다양한 병렬 처리 능력을 사용해 복잡한 컴퓨팅 문제를 해결하는 소프트웨어 개발을 가능케 하는 기술이다. 이 기술을 연산 최적화 알고리듬 중 PSO에 적용함으로써 알고리듬의 수행 속도를 개선하였다. CUDA를 적용한 PSO 알고리듬의 검증을 위해 언어 기반으로 프로그래밍하고 다양한 Test Function을 통해 시뮬레이션 하였다. 그리고 기존의 PSO 알고리듬과 비교 분석하였다. 또한 알고리듬의 성능 향상으로 여러 가지 최적화 분야에 적용 할 수 있음을 보인다.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.1
• /
• pp.67-72
• /
• 2000

Particularly based on the signal bias removal(SBR) algorithm for compensating the corrupted speech, this paper presents a new algorithm which is independent of environments, minimizes the amount of computation, and is readily applicable to the conventional recognition system. To this end, a multiple-bias algorithm and a partial codebook search algorithm have been added to the conventional SBR algorithm. The simulation results show that combining the two algorithms proposed in this paper provides a reduction of computation time to 1/8 times as well as an improvement of the recognition rate from 77.58% of the conventional system to 81.32%.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.4
• /
• pp.561-571
• /
• 1993

In this paper, a new discrete Symbol Timing Recovery (STR) algorithm, is proposed. This algorithm is derived from the optimum estimation theory. The algorithm combines the advantages of Mueller and $M\"{u}ller$ algorithm and Gardner algorithm, and avoids some of their shortcomings. The implementation of the new timing detector is simple and the combined operations of Carrier Recovery (CR) -STR is possible because the operation of the new STR is independent of the carrier phase. On the other hand, the behavior of nonlinear characteristics in the new algorithm is analyzed and explained. The performance evaluation is accomplished in detail by numerical calculations and Monte-Carlo simulations. In these respects, this algorithm is similar to Gardner's algorithm, but in tracking performance due to pattern jitter at small rolloff, the proposed algorithm is superior to Gardner's algorithm.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.5
• /
• pp.107-112
• /
• 1996

In this paper we propose a new fast codebook training algorithm for reducing the searching time of LBG algorithm. For each training data, the proposed algorithm stores the indexes of codewords that are close to that training data in the first iteration. It reduces computation time by searching only those codewords, the indexes of which are stored for each training data. Compared to one of the previous fast training algorithm, FSLBG, it obtains a better codebook with less exccution time. In our experiment, the performance of the codebook generated by the proposed algorithm in terms of peak signal-to-noise ratio(TSNR) is very close to that of LBG algorithm. However, the codewords to be searched for each training data of the proposed algorithm is only about 6%, for a codebook size of 256 and 1.6%, for a codebook size of 1.24, of LBG algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9C
• /
• pp.704-711
• /
• 2008

This paper proposes a new de-interlacing method which results in efficient visual improvement. In the proposed algorithm, the distance weight was considered and the previously developed the EDI (Edge Dependent Interpolation) algorithm and the NAL (New Adaptive Linear interpolation) algorithm were used as a basis. The do-interlacing method was divided into two main parts. First, the edge direction was found by using information of closer pixels. Then, missing pixels were interpolated along with the decided edge direction. In this paper, after predicting the edge through the EDI algorithm, missing pixels were interpolated by using the weighted distance based on the NAL algorithm. Experimental results indicate that the proposed algorithm be superior to the conventional algorithms in terms of the objective and subjective criteria.