Sequential Locally Optimum Test (SLOT): A Sequential Detection Scheme Based on Locally Optimum Test Statistic

Jinsoo BAE
Seong Ill PARK
Yun Hee KIM
Seokho YOON
Jongho OH
Iickho SONG
Seong-Jun OH

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E93-A    No.11    pp.2045-2056
Publication Date: 2010/11/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E93.A.2045
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Information Theory and Its Applications)
Category: Detection and Wireless Communications
asymptotic sample number,  locally optimum detector,  minimum false-alarm,  sequential test,  sequential probability ratio test,  

Full Text: PDF(867.3KB)>>
Buy this Article

Based on the characteristics of the thresholds of two detection schemes employing locally optimum test statistics, a sequential detection design procedure is proposed and analyzed. The proposed sequential test, called the sequential locally optimum test (SLOT), inherently provides finite stopping time (terminates with probability one within the finite horizon), and thereby avoids undesirable forced termination. The performance of the SLOT is compared with that of the fixed sample-size test, sequential probability ratio test (SPRT), truncated SPRT, and 2-SPRT. It is observed that the SLOT requires smaller average sample numbers than other schemes at most values of the normalized signal amplitude while maintaining the error performance close to the SPRT.