Detection of signals in nonstationary random noise VIA stationarization and stationarity test
Hiroshi Ijima, R. Okui, Akihiro Ohsumi · IEEE/SP 13th Workshop on Statistical Signal Processing, 2005 · 2005
An effective method is proposed for detecting a signal which is corrupted by nonstationary random noise. First, the noisy observation data whose statistical property is nonstationary owing to the noise is modified such that they become stationary ones. Then, based on this stationarized observation process, the stationarity test is executed to detect the signal. The method is tested by simulations to show the efficacy