OPTIMAL SAMPLING SCHEMES FOR THE GAUSSIAN HYPOTHESIS-TESTING PROBLEM

被引:20
作者
BAHR, RK [1 ]
BUCKLEW, JA [1 ]
机构
[1] UNIV WISCONSIN,DEPT ELECT & COMP ENGN,MADISON,WI 53706
来源
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING | 1990年 / 38卷 / 10期
关键词
6;
D O I
10.1109/29.60099
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
In this paper we discuss a class of sampling designs for a continuous time Gaussian hypothesis testing problem. It is a problem that arises frequently in systems where it is necessary to sample the continuous time signal to produce discrete time data for computer processing. We investigate a class of periodic nonuniform sampling designs which optimize performance under an asymptotic probability of error criterion. A design algorithm for the optimal sampler is presented. © 1990 IEEE
引用
收藏
页码:1677 / 1686
页数:10
相关论文
共 7 条
[1]   SAMPLING DESIGNS FOR THE DETECTION OF SIGNALS IN NOISE [J].
CAMBANIS, S ;
MASRY, E .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1983, 29 (01) :83-104
[2]  
CRAMER H, ACTUAL SCI IND, V736, P5
[3]  
CRAMER H, 1937, C CONSACRE THEORIE P, V3
[4]   LARGE DEVIATIONS FOR A GENERAL-CLASS OF RANDOM VECTORS [J].
ELLIS, RS .
ANNALS OF PROBABILITY, 1984, 12 (01) :1-12
[5]   ASYMPTOTIC EIGENVALUE DISTRIBUTION OF TOEPLITZ MATRICES [J].
GRAY, RM .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1972, 18 (06) :725-+
[6]  
Grenander U., 1958, TOEPLITZ FORMS THEIR
[7]  
RAO CR, 1973, LINEAR STATISTICAL I