tailieunhanh - Báo cáo hóa học: " Rapid Binary Gage Function to Extract a Pulsed Signal Buried in Noise"

Tuyển tập báo cáo các nghiên cứu khoa học quốc tế ngành hóa học dành cho các bạn yêu hóa học tham khảo đề tài: Rapid Binary Gage Function to Extract a Pulsed Signal Buried in Noise | EURASIP Journal on Applied Signal Processing 2004 13 1985-1992 2004 Hindawi Publishing Corporation Rapid Binary Gage Function to Extract a Pulsed Signal Buried in Noise Colin Ratcliffe Mechanical Engineering Department United States Naval Academy Annapolis MD 21402 USA Email ratcliff@ William J. Bagaria Aerospace Engineering Department United States Naval Academy Annapolis MD 21402 USA Email bagaria@ Sonia M. F. Garcia Mathematics Department United States Naval Academy Annapolis MD 21402 USA Email smg@ Richard P. Fahey Aerospace Engineering Department United States Naval Academy Annapolis MD 21402 USA Email rpfahey@ Received 1 August 2003 Revised 14 January 2004 Recommended for Publication by Sang Uk Lee The type of signal studied in this paper is a periodic pulse with the pulse length short compared to the period and the signal is buried in noise. If standard techniques such as the fast Fourier transform are used to study the signal the data record needs to be very long. Additionally there would be a very large number of calculations. The rapid binary gage function was developed to quickly determine the period of the signal and the start time of the first pulse in the data. Once these two parameters are determined the pulsed signal can be recovered using a standard data folding and adding technique. Keywords and phrases rapid binary gage function signal pulsed signal noise. 1. INTRODUCTION The motivation for this research was the need to extract pulsed signals that were buried in noise signal-to-noise ratio less than one . The signals of interest have a short pulse duration compared to the period of the signal and they are weak compared to the background noise. Another consideration when analyzing these signals was that the length of the recorded signal be as short as possible. Classic techniques for recovering information about noisy signals are usually based on the Fourier transform 1 the periodogram 2 or simply folding and adding