Testing of hypotheses and detection of known signals under the integrated power the integrated power criterion

적분검파력 결정 기준에서의 가설 검정과 알려진 신호 검파

  • 김선용 (한림대학교 전자공학과) ;
  • 송익호 (한국과학기술원 전기 및 전자공학과) ;
  • 장태주 (한국과학기술원 전기 및 전자공학과) ;
  • 김광순 (한국과학기술원 전기 및 전자공학과)
  • Published : 1996.03.01

Abstract

In this paper, a new test criterion for binary decision problems is proposed. The integrated power flunction over a parameter interval is first itroduced as an extension of the power function. The concept of the most integrated powerful (MIP) test based on the integrated power function is then introduced. The MIP criterion is to masimize the value of the integrated power function in any paricular parameter interval. As an applicationof the MIP test, the known signal detection problem is considered. The test statistic of the MIP detector for known signals is obtained and an approximation to the MIP test statistic is also considered.

Keywords

References

  1. An Introduction to Probability Theoryand Mathematical Statistics V. K. Rohatgi
  2. Signal Detection in Non-Gaussian Noise S. A. Kassam
  3. IEEE Tr. Inform. Theory v.IT-38 Locally optimum rand detection of correlatedrandom signals in additive noise I. Song;S. A. Kassam
  4. 한국음향학회 논문지 v.12 신호의존성 잡음 모형과 복합신호검파 송익호;김상엽;김선용;박성일
  5. IEEE Tr. Inform. Theory v.IT-41 On the score functions of the two-sample locallyoptiumu rank test statistic for random signals in additive noise S. Y. Kim;I. Song
  6. Proc. Inter. Conf. Acoust. Speech, Signal Processing v.4 Performance of various detectorsin non-Gaussian noise and weak stochastic signal environment L. C. Godara;M. Schultheiss
  7. Mathematical Statistics:A Decistion Theoretic Approach, Academic T. S. Ferguson