Title :
Optimal Design of Noise-Enhanced Binary Threshold Detector Under AUC Measure
Author :
Gencheng Guo ; Xinwei Yu ; Yindi Jing ; Mandal, Mrinal
Author_Institution :
Electron. & Inf. Eng. Coll., Henan Univ. of Sci. & Technol., Luoyang, China
Abstract :
This letter considers the binary threshold system (TS) based detector for a general binary testing problem. First, the optimal binary TS that maximizes the area under the ROC curve (AUC), where ROC stands for the receiver operating characteristic, is derived. Then the noise-enhanced effect is investigated. The optimal noise that can achieve the maximum AUC is derived and shown to be deterministic. An example is shown to help justify the derived results.
Keywords :
sensitivity analysis; signal detection; AUC measure; ROC curve; binary threshold system; general binary testing problem; noise-enhanced binary threshold detector; noise-enhanced effect; optimal noise; receiver operating characteristic; Detectors; Educational institutions; Noise; Noise measurement; Probability density function; Robustness; Signal detection; AUC; noise-enhanced effect; threshold detector; threshold system;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2012.2234452