• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.7
• /
• pp.66-77
• /
• 1997

This paper presents an automatic tuning method for parameters of a multivaiable self-tuning velocity-type PID controller which adapts to changes in the system parameters with time delays and noises. The velocity-type PID control structure is determined in the process of minimizing the variance of the auxiliarly output, and self-tuning effect is achieved through the recursive least square algorithm at the parameter estimation stage and also through the Robbins-Monro algorithm at the stage of optiminzing the design parameters of the controller. The proposed PID type multivariable self-tuning method is simple andeffective compared with other esisting multivariable self-tuning methods. Computer simulation has shown that the proposed algorithm is beter than the trial-and-error method in the tracking performance.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.497-500
• /
• 2005

This paper presents a design of a hybrid sensor chip which integrates an A/D converter module and a phase measurement module for measuring power line phase. Recursive sliding DFT based phase measurement module is designed using time shared multiplier which can reduce the size of SoC implementation. A/D converter is based on the sigma delta modulation in order to minimize the implementation space of the analog part and designed to obtain 8-bit resolution. Computer simulations and FPGA implementation are performed to verify hybrid sensor chip design. The hybrid sensor chip for 4-channel power line phase measurement is fabricated by using 0.35 micrometer CMOS process.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.1007-1010
• /
• 2005

This paper presents a fingerprint image registration method. In the fingerprint system, the insufficiency of mutual information between a template and a query fingerprint is one of major factors to deteriorate recognition performance. To overcome this problem, we need to register multiple impressions and integrate their information. Our method matches the ridges from multiple impressions recursively and then registers the impressions to minimize the registration error calculated from the Distance map. Our method use regularized TPS model as the transformation model to compensate for the plastic deformation. We compare our method with 3 prior arts (ICP, Distance Map, Ross's method). Our registration error and its' variance is the smallest and also the average registration error is below 3 pixels.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.3-6
• /
• 2005

Mobile Network의 이동성에 대한 연구가 활발하게 진행되어 오면서, 이동네트워크의 복잡한 모델인 Nested Mobile Network에 대한 연구가 부각되고 있다. 이 Nested Mobile Network에 대한 연구에는 네트워크 내에 존재하는 노드와 외부의 CN(Correspondent Node)과의 패킷전송에서 경로최적화를 하기 위한 방법 등에 대한 연구가 진행되어 오고 있으며, Nested Mobile Network에서 Pinball routing 문제 등으로 인해 경로최적화가 이루어지지 못하고, 이 문제가 일반적인 Nested Mobile Network에서 패킷사이즈의 길이를 지나치게 길어지게 함으로써 전송지연을 발생시키는 것이다, 본 논문에서는 반복적인 CoA(Care-of-Address)의 치환과정을 통해 Nested Mobile Network 상의 Pinball routing 문제를 해결하기 위한 효율적인 방안을 제안하고자 한다.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.21 no.3
• /
• pp.1-5
• /
• 1984

It is investigated that Mahalanobis distances are invariant under a proper transformation and interest MDs are distributed as non-central chi-square while it is well known that intraset MDs are distributed as central chi-square. And their moments have been expressed as simple polynomials whose coefficients satisfy straightforward recursive relations.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.12 no.2
• /
• pp.203-211
• /
• 2012

This paper proposes a $16{\times}16$ and $32{\times}32$ inverse transform architecture for HEVC (High Efficiency Video Coding). HEVC large transform of $16{\times}16$ and $32{\times}32$ suffers from huge computational complexity. To resolve this problem, we proposed a new large inverse transform architecture based on hardware reuse. The processing element is optimized by exploiting fully recursive and regular butterfly structure. To achieve low area, the processing element is implemented by shifters and adders without multiplier. Implementation of the proposed 2-D inverse transform architecture in 0.18 ${\mu}m$ technology shows about 300 MHz frequency and 287 Kgates area, which can process 4K ($3840{\times}2160$)@ 30 fps image.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.6
• /
• pp.350-356
• /
• 2013

Single Carrier Frequency Domain Multiple Access (SC-FDMA) has proven to be the best long term evolution for uplink multiple access because of its low Peak to Average Power Ratio (PAPR), a feature that leads to low power consumption. This is achievable only if the resource allocation is performed in a contiguous manner. This paper proposes a new approach with an improvement in the global resources allocation. The new approach presented utilizes the gain function, which adopts some of the procedures deduced from the older Recursive Maximum Expansion (RME) algorithm. The experiment proved that the new approach is better than the original RME algorithms and in most cases, is closer to the optimal solution.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.1
• /
• pp.17-23
• /
• 2020

We propose a system time synchronization method between embedded Linux-based distributed control devices by using Transmission Control Protocol (TCP) communication and General Purpose Input Output (GPIO) device. The GPIO signal is used as the trigger signal for synchronization and the TCP communication is used to transfer the system time of master Linux, which serves as the reference clock, to slave Linux. Precise synchronization performance is achieved by measuring and compensating for the propagation delay of GPIO signal and the acquisition and setting latency of Linux system time. We build an experimental setup consisting of two embedded Linux systems, and perform extensive experiments to verify the performance of the proposed synchronization method.

• Journal of Industrial Technology
• /
• v.10
• /
• pp.15-20
• /
• 1990

A parallel processing algorithm for discrete Kalman filter, which is one of the most commonly used filtering technique in modern control, signal processing, and communication, is proposed. Previously proposed parallel algorithms to decrease the number of computations needed in the Kalman filter are the hierachical structures by distributed processing of measurements, or the systolic structures to disperse the computational burden. In this paper, a new parallel Kalman filter employing a structure similar to recursive doubling is proposed. Estimated values of state variables by the new algorithm converge with two times faster data processing speed than that of the conventional Kalman filter. Moreover it maintains the optimality of the conventional Kalman filter.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1993.06a
• /
• pp.75-79
• /
• 1993

본 논문에서는 변화가 심한 음원파형을 추정하는 새로운 순차처리 알고리듬을 제안한다. 먼저, 1) 기존의 순차처리 분석법중 대표적인 분석법인 RLS(recursive least square)의 문제점들을 검토하고, 2) 이를 개선하기 위해서 관측행렬(observation matrix)을 최적차수의 SVD(reduced-rank singular value decomposition)로 재구성하고, 3) 이에 견실개념(robustness concept)을 적용해서 최적의 성도변수(vocal tract parameter)를 찾아내고 역필터를 적용해서 음원(voice source)을 효과적으로 구분해낸다. 본 논문에서 제안된 방법으로 음원을 추정할 경우, 변화가 심한 음원파형을 잘 추정할 수 있으며, 음원의 특성을 구분해낸 성도 파라미터도 효과적으로 추정할 수 있다. 본 연구내용은 음성합성에서 자연성 개선 및 개인성 구현을 위해서 필수적이며, 다양한 형태의 음성을 표현하기 위해 사용되어질 수 있다. 또한, 음성코딩, 화자인식, 음성인식에서도 사용되어질 수 있다.