• Title/Summary/Keyword: Error Reduction

Search Result 1,416, Processing Time 0.026 seconds

Non-linear Data Classification Using Partial Least Square and Residual Compensator (부분 최소 자승법과 잔차 보상기를 이용한 비선형 데이터 분류)

  • 김경훈;김태영;최원호
    • Journal of Institute of Control, Robotics and Systems
    • /
    • v.10 no.2
    • /
    • pp.185-191
    • /
    • 2004
  • Partial least squares(PLS) is one of multiplicate statistical process methods and has been developed in various algorithms with the characteristics of principal component analysis, dimensionality reduction, and analysis of the relationship between input variables and output variables. But it has been limited somewhat by their dependency on linear mathematics. The algorithm is proposed to classify for the non-linear data using PLS and the residual compensator(RC) based on radial basis function network (RBFN). It compensates for the error of the non-linear data using the RC based on RBFN. The experimental result is given to verify its efficiency compared with those of previous works.

Channel Compensation for Cepstrum-Based Detection of Laryngeal Diseases (켑스트럼 기반의 후두암 감별을 위한 채널보상)

  • Kim Young Kuk;Kim Su Mi;Kim Hyung Soon;Wang Soo-Geun;Jo Cheol-Woo;Yang Byung-Gon
    • MALSORI
    • /
    • no.50
    • /
    • pp.111-122
    • /
    • 2004
  • Automatic detection of laryngeal diseases by voice is attractive because of its non-intrusive nature. Cepstrum based approach to detect laryngeal cancer shows reliable performance even when the periodicity of voice signals is severely lost, but it has a drawback that it is not robust to channel mismatch due to different microphone characteristics. In this paper, to deal with mismatched training and test microphone conditions, we investigate channel compensation techniques such as Cepstral Mean Subtraction (CMS) and Pole Filtered CMS (PFCMS). According to our experiments, PFCMS yields better performance than CMS. By using PFCMS, we obtained 12% and 40% error reduction over baseline and CMS, respectively.

  • PDF

A Train Ticket Reservation Aid System Using Automated Call Routing Technology Based on Speech Recognition (음성인식을 이용한 자동 호 분류 철도 예약 시스템)

  • Shim Yu-Jin;Kim Jae-In;Koo Myung-Wan
    • MALSORI
    • /
    • no.52
    • /
    • pp.161-169
    • /
    • 2004
  • This paper describes the automated call routing for train ticket reservation aid system based on speech recognition. We focus on the task of automatically routing telephone calls based on user's fluently spoken response instead of touch tone menus in an interactive voice response system. Vector-based call routing algorithm is investigated and mapping table for key term is suggested. Korail database collected by KT is used for call routing experiment. We evaluate call-classification experiments for transcribed text from Korail database. In case of small training data, an average call routing error reduction rate of 14% is observed when mapping table is used.

  • PDF

Efficient Data Reduction for Point-Based Rendering using Extended QEM (효율적인 점 기반 렌더링을 위한 확장 이차 오류 척도 기반의 간략화 방법 개발)

  • Kim Duck-bong;Kang Eui-chul;Lee Kwan H.;Pajarola Renato B.
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2005.11a
    • /
    • pp.712-714
    • /
    • 2005
  • 본 논문은 효율적인 점 기반 렌더링(Point-based Rendering)을 위해 확장 이차 오류 척도(Quadric Error Metrics) 기법을 이용하는 간략화 알고리즘을 제안한다. 점 기반 렌더링의 기본 개념은 자유곡면을 메쉬와 같은 연결정보 없이 직접 점들로 표현하고, 렌더링하는 것이다. 확장 이차 오류 척도 기법은 메쉬를 간략화 하는데 있어 기하 정보뿐만 아니라 색상, 텍스쳐 좌표 정보까지 고려하여 간략화 하는 알고리즘이다. 이 연구는 3차원 점 데이터로부터 복원한 폴리곤 메쉬 모델로부터 효율적인 점 기반 렌더링(Point-based Rendering)을 위해 기하 정보 및 색상 정보까지 고려하여 원본 점 데이터를 간략화 하는 저용량의 효율적인 점 기반 렌더링 알고리즘을 제안하고, GPU 기반 렌더링 결과를 보였다.

  • PDF

Design of a Cost-Effective Hybrid-Type PBEx Providing a High Power Budget in an Asymmetric 10G-EPON

  • Kim, Kwangok;Lee, Sangsoo;Lee, Jonghyun;Jang, Younseon
    • ETRI Journal
    • /
    • v.34 no.6
    • /
    • pp.838-846
    • /
    • 2012
  • This paper proposes a cost-effective hybrid-type power budget extender (PBEx) that can provide a high power budget of over 45 dB in an asymmetric 10-Gb/s Ethernet passive optical network (10/1G-EPON). The hybrid-type 10/1G-EPON PBEx comprises a central office terminal (COT) and remote terminal (RT) module supporting four channels and uses a coarse wavelength division multiplexing (CWDM) technology between the COT and RT for a reduction of fiber cost and efficient access network design. The proposed 10/1G-EPON PBEx can provide over a 40-km reach and 128-way split per CWDM wavelength with no modification of a legacy 10/1G-EPON system and can satisfy the error-free service in $10^{10}$ packet transmission.

Replica Correlation-Based Synchronization with Low Complexity and Frequency Offset Immunity

  • Chang, Kapseok;Bang, Seung Chan;Kim, Hoon
    • ETRI Journal
    • /
    • v.35 no.5
    • /
    • pp.739-747
    • /
    • 2013
  • This paper investigates the multifarious nature of the long-term evolution (LTE) scheme and that of the modified LTE scheme for symbol timing synchronization (STS). This investigation allows us to propose a new replica correlation-based STS scheme to overcome the inherent weaknesses of the other two schemes. The proposed STS signal combines a gold sequence and a half sine wave in the time domain, whereas conventional STS signals specify either binary sequences or complex sequences in the time domain or in the frequency domain. In the proposed scheme, a sufficient correlation property is realized by the gold sequence, and robustness against the frequency offset (FO) is achieved through the sine wave. Compared to the existing LTE-related schemes, the proposed scheme can better achieve immunity to FO and reduction in detector complexity, as well as a low peak-to-average power ratio and a low detection error rate. Performance evaluations through analysis and simulation are provided in the paper to demonstrate these attributes.

Application of process monitoring with reduced order model and EKF to distillation column (차수 감소 모델과 EKF를 이용한 공정 모니터링의 응용)

  • 김태민;양대륙
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1997.10a
    • /
    • pp.1766-1769
    • /
    • 1997
  • Fast and accurate distillation design requires a model that significantly reduces the problem size withour loss of accruacy is especially suitable for rela-time applicatoins. the reduced order model is obtained by use of Principal Component Anlysis(PCA). Then the extended Kalman filter and the Recursie Predictiuon Error(RPE) mehtod are appliced to identify the model parameters and the feed compostion form the measuremenets of the coumn. as a consequence it is found that the model reduction thechique can account for the dynamics of the rigorous distillation model and not only the model parameters, bu also the feed compostion can be identified efficiently. this technique is applied to industrial operation data verify the performance of reduced order model.

  • PDF

Error Reduction of Sliding Mode Control Using Sigmoid-Type Nonlinear Interpolation in the Boundary Layer

  • Kim, Yoo-K.;Jeon, Gi-J.
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 2003.10a
    • /
    • pp.1810-1815
    • /
    • 2003
  • Sliding mode control with nonlinear interpolation in the boundary layer is proposed. A modified sigmoid function is used for nonlinear interpolation in the boundary layer and its parameter is tuned by a fuzzy logic controller. The fuzzy logic controller that takes the distance between the system state and the sliding surface as its input guides the choice of parameter of the modified sigmoid function and the parameter is on-line tuned. Owing to the decreased thickness, the proposed method has better tracking performance than the conventional linear interpolation method. To demonstrate its performance, the proposed control algorithm is applied to a simple nonlinear system model.

  • PDF

Design of Carrier Recovery Loop for QPSK Demodulator (QPSK 복조기를 위한 반송파 복구 회로 설계)

  • 하창우;김형균;김환용
    • Proceedings of the IEEK Conference
    • /
    • 2000.11a
    • /
    • pp.85-88
    • /
    • 2000
  • In order to resolve problems according to the phase error in QPSK demodulator of the digital communication systems. The demodulator requires carrier recovery loop which searches for the frequency and phase of the carrier. In this paper the complexity of implementation is reduced by the reduction into half of the number of the multiplier in filter structure of the conventional carrier recovery loop, and as the drawback of NCO of the conventional carrier recovery loop wastes a amount of power for the structure of lookup table , We designed the structure of combinational logic without the lookup table. In the comparison with dynamic power of the proposed NCO, the power of NCO with the lookup table is 175㎼, NCO with the proposed structure is 24.65㎼. As the result, it is recognized that about one eight of loss power is reduced. In the simulation of carrier recovery loop designed QPSK demodulator, it is known that the carrier phase is compensated.

  • PDF

(Suboptimal Detection Thresholds for Tracking in Clutter) (클러터 환경에서의 표적 추적을 위한 준최적의 검출 문턱값)

  • Jeong, Yeong-Heon;Sin, Han-Seop
    • Journal of the Institute of Electronics Engineers of Korea SC
    • /
    • v.39 no.2
    • /
    • pp.176-181
    • /
    • 2002
  • In this paper, we consider the optimal control of detection threshold to minimize the conditional expectation of mean-square state estimation error for a probabilistic data association (PDA) filter. Earlier works on this problem involved the cumbersome graphical optimization algorithm or time-consuming numerical optimization algorithm. Using the numerical approximation of information reduction factor, we obtained the suboptimal detection threshold in a closed-form. This results are very useful for real- time implementation.